An Evidence-Based Safety Management System for Heavy Truck Transport Operations

An Evidence-based Safety Management System

for Heavy Truck Transport Operations

Lori Mooren

A thesis in fulfilment of the requirements for the degree of

Doctor of Philosophy

School of Aviation

Faculty of Science

June, 2016 THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet Surname or Family name: Mooren First name: Lori Other name/s: Elyse Abbreviation for degree as given in the University calendar: PhD School: School of Aviation: Faculty of Science Title: An evidence-based safety management system for heavy truck transport operations

Abstract

Drawing from the knowledge from prior occupational safety and road safety research (Study 1). a study of safety management characteristics comparing those in good safety performing heavy vehicle operators and poor safety performers sought to synthesise the distinguishing features between them. Two empirical studies were conducted (Studies 2 and 3). The first was a survey of senior managers of Australian heavy vehicle operating companies. The second was an in-depth investigation of a sample of the survey participants to validate the self-reported survey, and to learn more about the reported characteristics and non-reported characteristics in situ. The findings of these studies provided the basis upon which to build a safety management system (SMS) suitable for heavy transport vehicle operations. This process resulted in the identification of 14 safety manag ement characteristics that have strong research evidence for inclusion in a safety management system (SMS) for heavy truck operations. These findings, together with analysis of sound theoretical models to underpin the SMS, were used to shape the SMS.

The SMS features three spheres of management practices - risk assessment and management. driver risk management and safety culture management. Drawing from the li terature. a dynamic model of a safety management system is presented and explained. The original aim of this thesis research has been met, providing an evidence-based safety management system that is likely to reduce crash and injury risk when applied to heavy vehicle transport operations.

Declaration relating to disposition of project thesis/dissertation I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known. subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or disserta tion. I also authorise University Microfilms to use the 337-word abstract of my thesis in Dissertation Abstracts International (this is aoolicable to doctorallheses onlv\.

The University recog nises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and requ ire the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE TH ESIS

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed ……………………………………………......

Date ……………………………………………......

AUTHENTICITY STATEMENT

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Signed ……………………………………………......

Date ……………………………………………......

ORIGINALITY STATEMENT

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ……………………………………………......

Date ……………………………………………...... Table of Contents

Originality Statement ...... vii

Prefacing Comments and Declarations regarding Publications used in this Thesis ...... viii

Acknowledgements ...... ix

Abbreviations & Glossary ...... x

List of Tables ...... xi

List of Figures ...... xiv

Extended Abstract ...... 1

Chapter 1: Background and the Australian Trucking Industry ...... 2 1.1 Background and introduction ...... 2 1.2 The size of the safety problem in the road freight transport industry ...... 3 1.3 The dimensions of the problem ...... 5 1.4 Road freight transport industry structure and inherent pressures on safety ...... 6 1.4.1 Basis of risk in the truck driving occupation ...... 6 1.4.2 Road freight transport industry structure and inherent safety risks ...... 11 1.5 Safety regulation of the road freight transport industry ...... 18 1.6 Compliance and alternative compliance schemes ...... 21 1.6.1 Alternative compliance or ‘government concessional’ schemes ...... 22 1.6.2 Industry safety management accreditation schemes ...... 23 1.6.3 Relationships between accreditation and safety outcomes ...... 25 1.7 Constraints on regulatory effectiveness ...... 27 1.8 Conclusions ...... 27 1.9 Thesis structure and content ...... 29

Chapter 2: Current Approaches to Safety Management ...... 30 2.1 What is safety management? ...... 30 2.2 Road safety history and the development of scientific approaches ...... 30 2.3 Applying road safety approaches to work related driving ...... 36 2.4 Safety management approaches in other sectors ...... 40 2.4.1 Models of WHS accident factor analysis ...... 41 2.4.2 Identifying weaknesses in organisational safety defences through accident analysis ...... 47 2.5 What is a safety management system? ...... 49

ii 2.6 Applicability of SMS to the road freight transport sector ...... 51 2.7 Conclusions ...... 52

Chapter 3: Strategic Scientific Literature Review (Study 1) ...... 55 3.1 Introduction and aims of this Study 1: systematic review of scientific literature ...... 55 3.2 Literature review methods ...... 56 3.3 Results of the strategic literature review (Study 1) ...... 57 3.3.1 Studies of organisational characteristics and safety results at organisational level ...... 58 3.3.2 Studies of organisational or personal characteristics and safety results at employee level ...... 64 3.3.3 Studies of the effects of organisational interventions on safety outcomes ...... 71 3.4 Summary of results ...... 74 3.5 Discussion of literature findings ...... 77 3.6 Conclusions of the strategic literature analysis (Study 1) ...... 82

Chapter 4: Survey of Characteristics Distinguishing Between Better and Poorer Safety Performers (Study 2) ...... 85 4.1 Introduction ...... 85 4.2 Methods for Study 2 ...... 86 4.2.1 Design ...... 86 4.2.2 Participants ...... 87 4.2.3 Materials ...... 87 4.2.4 Procedure ...... 89 4.2.5 Response rate ...... 90 4.2.6 Survey analysis method ...... 90 4.3 Results of the manager survey (Study 2) ...... 91 4.3.1 Freight and vehicle fleet ...... 91 4.3.2 Journey and risk assessment ...... 92 4.3.3 Staffing and driver recruitment ...... 93 4.3.4 Pay and conditions ...... 94 4.3.5 Policies and safety accreditation ...... 95 4.3.6 Scheduling and training ...... 96 4.3.7 Communication and driver participation in workplace health & safety ...... 97 4.3.8 Work monitoring ...... 98 4.3.9 Driver discipline and incentives ...... 99 4.3.10 Incidents and record keeping ...... 99 4.4 Discussion of the manager survey findings ...... 100 4.4.1 Limitations of Study 2 ...... 103

iii 4.5 Conclusions of the manager survey (Study 2) ...... 104

Chapter 5: In-depth Investigation Study (Study 3) ...... 106 5.1. Introduction ...... 106 5.1.1 Specific Study 2 findings ...... 106 5.1.2 Case for validating Study 2 survey findings (Study 3) ...... 107 5.2. Methods ...... 108 5.2.1 Design ...... 108 5.2.2 Participants ...... 109 5.2.3 Measures and procedures ...... 110 5.2.4 Analysis method ...... 112 5.3 Results of the in-depth investigation (Study 3) ...... 113 5.3.1 Truck management characteristics ...... 114 5.3.2 Scheduling, journey and site risk assessment and response to drivers’ safety concerns ...... 118 5.3.3 Driver employment and remuneration ...... 123 5.3.4 Policies, accreditations, and key performance indicators (KPIs) for safety management ...... 128 5.3.5 Driver input into WHS decisions and safety training ...... 133 5.3.6 Monitoring, discipline and incentives ...... 138 5.4 Differences in consistencies between lower and higher claimers ...... 143 5.5 Additional interview questions ...... 144 5.6 Discussion ...... 146 5.6.1 Validated safety management characteristics (27) ...... 147 5.6.2 Safety management characteristics not validated in Study 3 (4) ...... 151 5.6.3 Inconclusive safety management characteristics (6) ...... 152 5.6.4 Safety cultural characteristics ...... 153 5.6.5 Limitations of the in-depth investigation ...... 154 5.7 Conclusions of the in-depth investigation (Study 3) ...... 155 5.7.1 Implications and next steps ...... 156

Chapter 6: Discussion and Conclusions – Evidence-based Safety Management System (SMS) ...... 157 6.1 Background to this thesis ...... 157 6.2 Aims of this thesis ...... 158 6.3 Approach taken in this thesis research ...... 158 6.4 Summary of Study 1, 2 and 3 findings ...... 159 6.4.1 Strategic literature review (Study 1) findings ...... 159 6.4.2 Survey of managers (Study 2) findings ...... 160 6.4.3 In-depth investigation validation of survey (Study 3) findings ...... 160

iv 6.5 Set of evidence-based safety management characteristics ...... 162 6.5.1 Evidence-based safety risk assessment and management practices (6) ...... 164 6.5.2 Evidence-based driver risk management practices (6) ...... 167 6.5.3 Evidence-based safety culture management practices (2) ...... 170 6.6 How the evidence-based safety management practices combine and interact ...... 172 6.7 General discussion ...... 175 6.7.1 Limitations ...... 176 6.7.2 Empirical testing of the SMS ...... 176 6.7.3 Recommendations for further research ...... 177 6.7.4 Implications for industry, regulators and insurers ...... 178 6.7.5 Changing the paradigm ...... 178 6.8 Conclusions ...... 180

References ...... 182

Appendix A – Ten-Point National Logistics Safety Code ...... 199

Appendix B - Manager Survey Questionnaire ...... 200

Appendix C - Recording sheet for in-depth investigation ...... 208

Appendix D - Driver Survey Questionnaire ...... 212

v Acknowledgements

An initial grant from the NSW Motor Accidents Authority provided funding for a fellowship for the author to develop a body of research to address heavy vehicle transport safety beyond the regulatory systems in place. Later, funding and other support that made this particular study possible was provided by the ARC Linkage Grant LP100100283 in partnership, the NSW Centre for Road Safety, Transport for NSW, Transport Certification Australia, National Transport Commission, Zurich Financial Services, and the Motor Accidents Authority of NSW. I especially want to thank Partner Organisation representatives, Dr Soames Job and Greg Dikranian of Transport for NSW, Peter Johansson and Roger Hancock of Zurich, Dr Charles Karl and Gavin Hill of Transport Certification Australia, Christine Baird, Carmel Donnelly and the late Pam Albany of the Motor Accidents Authority and Dr Jeff Potter of the National Transport Commission for their helpful support and advice throughout the project.

I am grateful to the participating company representatives who generously gave their time to provide the needed data. Helpful comments on the draft questionnaire for Study 2 were provided by Peter Elliot of the Australian Logistics Council, Robert Howse of NatRoads, Owen Driscoll of National Transport Insurance, Alan Bettison of TNT, Merry Manton of Simon National Carriers, Justin Fleming of the Australian Trucking Association, and Dr Junjira Mahaboon, former PhD scholar at Transport and Road Safety (TARS).

I am extremely indebted to Dr Rena Friswell for her guidance and assistance throughout this project. Faisal Magableh is also gratefully acknowledged for assisting with recruitment, conducting some of the interviews and coding and entering data.

Professors Ann Williamson and Raphael Grzebieta and Associate Professor Jake Olivier, as Chief Investigators on the project, provided superior technical guidance throughout the project as well as their motivational support. Beyond this, as my PhD supervisors I am endlessly grateful to Professors Ann Williamson and Raphael Grzebieta for their kind patience and steadfast determination to assist me to complete this thesis.

My dear friend, Dr. Wendy Sarkissian, generously provided invaluable editorial advice on this work, for which I am very grateful.

Finally, I express my appreciation to my partner, Emeritus Professor David Wilmoth, for his continued encouragement and support to complete my PhD work.

ix Prefacing Comments and Declarations regarding Publi cations used in this Thesis

The research that formed the basis of this thesis involved a joint effort by a team of UNSW researchers. My supervisors, Professors Ann Williamson and Raphael Grzebieta were two of the Chief Investigators on the project funded by an Australian Research Council Linkage grant. Associate Professor Jake Olivier was the third Chief Investigator. Dr Rena Friswell was also a member of the team and contributed significantly to the project. Chapters 3 and 4 were largely drawn from papers co-authored by all members of the project team. These are: Mooren, L., Grzebieta, R., Williamson, A., Olivier, J., & Friswell, R. (2014). Safety management for heavy vehicle transport: A review of the literature. Safety Science, 62(0), 79-89. doi:http:/ldx.doi.org/10.1016 /j.ssci.2013.08.001 Mooren, L., Williamson, A. , Friswell, R., Olivier, J., Grzebieta, R., &Magableh , F. (2014). What are the differences in management characteristics of heavy vehicle operators with high insurance claims versus low insurance claims? Safety Science, 70(0), 327-338. doi:http:/ldx.doi.org/10.1 016{j.ssci.2014.07.007

The publication contributions to all papers relating to this thesis are estimated in the table below.

Author/investigator % contribution Nature of contribution Lori Mooren 70 Designed and conducted the research, analysed data, drafted papers Ann Williamson 10 Advised and assisted study design, analysis, interpretation, and extensive editing of papers Rena Friswell 10 Advised and assisted study design, analysis and interpretat,ion and assisted writing of papers Raphael Grzebieta 3 Advised on study methods and papers Jake Olivier 5 Advised on statistical methods and interpretation and papers Faisal Magableh 2 Assisted with the survey and data analysis

Candidate Declaration

I certify that the parts of the co-authored publications were a result of my research towards this PhD, and that reproduction in this thesis does not breach copyright regulations.

Lori Mooren (Candidate) Date

viii Abbreviations & Glossary

ABS Anti-lock braking system ALC Australian Logistics Council ATA Australian Trucking Association ATC Australian Transport Council – State and Federal Transport Ministers ATSB Australian Transport Safety Bureau BFM Basic Fatigue Management BITRE Australian Bureau of Infrastructure, Transport and Regional Economics CoR Chain of Responsibility Defect notice Document issued by police or road authority about a vehicle fault Eco driving A smooth driving style to achieve low fuel usage ESP Electronic stability program FMCSA (US) Federal Motor Carrier Safety Administration GMV Gross Vehicle Mass GPS Global positioning system HAZMAT Hazardous materials HV Heavy vehicle HVNL (Australian) Heavy Vehicle National Law KPI Key performance indicator License points Demerit points as penalties for traffic offences N Frequency N Number in sample NHVAS National Heavy Vehicle Accreditation Scheme NHVR (Australian) National Heavy Vehicle Regulator NSW New South Wales NTC (Australian) National Transport Commission OHS Occupational health and safety (now more commonly WHS) Optalert Special glasses that detect eye closures used while driving OR Odds ratio OSC Organisational safety climate RSRT (Australian) Road Safety Remunerations Tribunal Safety-cams Fixed highway cameras used to measure time and distance travelled by vehicles SD Standard deviation SMS Safety management system Toolbox talks Employee meetings that provide opportunities for discussing issues of concern TruckSafe An Australian industry-managed safety management accreditation scheme TWUA Transport Workers Union Australia Underrun device A vehicle feature that prevents smaller vehicles going under a truck US United States of America UNSW The University of New South Wales Australia WAHVAS Western Australia Heavy Vehicle Accreditation Scheme WHS Workplace health and safety Yellow Pages The telephone book listing Australian businesses

x List of Tables

Table 1.1 Effects of Driver Payment Methods on Risk Behaviour ...... 8 Table 1.2. Effects of Driver Pay Levels on Safety Behaviour and Outcomes ...... 10 Table 3.1 Studies of safety factors and outcomes with company as the unit of analysis ...... 60 Table 3.2 Studies of safety factors and outcomes with the individual as the unit of analysis...... 66 Table 3.3 Studies of organisational interventions and their effects on safety outcomes ...... 72 Table 3.4 Summary of the number of studies showing significant relationships between the characteristic and safety outcomes for organisation level, individual level and intervention studies ...... 75 Table 3.5 Summary of the number of heavy vehicle transport safety studies showing significant relationships between the characteristic and safety outcomes for organisation level, individual level and intervention studies ...... 77 Table 4.1. Survey response rates and number eligible for the study using criteria established from insurance criteria for those who completed the survey questionnaire using both methods of recruitment ...... 90 Table 4.2 Considerations involved in truck purchasing decisions by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers ...... 92 Table 4.3 Journey and site risk assessment by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than low claimers...... 93 Table 4.4 Driver staffing and recruitment practices for companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers ... 94 Table 4.5 Driver payment practices for companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers ...... 95 Table 4.6 Safety policies and safety accreditation by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers ... 96 Table 4.7 Scheduling and training by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers ...... 97 Table 4.8 Communication and driver input for companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers ...... 98 Table 4.9 In-vehicle monitoring by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers ...... 98 Table 4.10 Driver discipline practices and safety incentives by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers ...... 99

xi Table 4.11 Summary of expected and unexpected findings comparing 37 safety characteristics of low and higher insurance claimers where higher odds ratio indicates higher claimers have larger odds than lower claimers ...... 100 Table 5.1 Types of data collected in the original survey and in the in-depth investigations ...... 111 Table 5.2 Consistency for truck management characteristics – numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent...... 116 Table 5.3 Manager-provided evidence regarding trucks – percentages of companies showing consistency with their responses in the original survey, through managers’ interview responses and observations/documentation...... 117 Table 5.4 Driver-provided evidence regarding trucks – percentages of companies showing consistency with their managers’ responses in the original survey through drivers’ interview responses. .... 117 Table 5.5 Consistency for scheduling, risk assessment and responding to drivers’ concerns – numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in- depth study. Bolded fractions denote half or less were consistent...... 120 Table 5.6 Manager-provided evidence regarding scheduling, risk assessment and responding to drivers’ concerns – percentages of companies showing consistency with their responses in the original Survey through managers’ interview responses and observations/documentation...... 121 Table 5.7 Driver-provided evidence regarding scheduling, risk assessment and responding to drivers’ concerns – percentages of companies showing consistency with the responses by their managers in the original Survey through drivers’ interview responses...... 122 Table 5.8 Consistency for employment and remuneration characteristics – numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent...... 125 Table 5.9 Manager-provided evidence regarding employment and remuneration – percentages of companies showing consistency with their responses in the original Survey through managers’ interview responses and observations/documentation...... 126 Table 5.10 Driver-provided evidence regarding employment and remuneration – percentages of companies showing consistency with the responses by their managers in the original Survey through drivers’ interview responses...... 127

xii Table 5.11 Consistency for policies, accreditations and KPI characteristics – numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent...... 130 Table 5.12 Manager-provided evidence regarding policies, accreditations and KPIs – percentages of companies confirming or not confirming their responses in the original Survey through managers’ interview responses and observations/documentation...... 131 Table 5.13 Driver-provided evidence regarding policies, accreditations and KPIs – percentages of companies showing consistency with the responses by their managers in the original Survey through drivers’ interview responses...... 132 Table 5.14 Consistency for driver input and safety training characteristics – numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent...... 135 Table 5.15 Manager-provided evidence regarding driver input into WHS and safety training – percentages of companies showing consistency with their responses in the original Survey through managers’ interview responses and observations/documentation...... 136 Table 5.16 Driver-provided evidence regarding driver input into WHS and safety training – percentages of companies showing consistency with the responses by their managers in the original Survey through drivers’ interview responses...... 137 Table 5.17 Consistency for monitoring, discipline and incentives characteristics – numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent...... 140 Table 5.18 Manager-provided evidence regarding driver monitoring, discipline and incentives – percentages of companies showing consistency with their responses in the original Survey through managers’ interview responses and observations/documentation...... 141 Table 5.19 Driver-provided evidence regarding driver monitoring, discipline and incentives – percentages of companies showing consistency with the responses by their managers in the original Survey through drivers’ interview responses...... 142 Table 5.20 Summary of 37 safety management practices confirmed (Y) or not (N) or not tested (-) by the in-depth investigation study compared with Study 2 findings ...... 144 Table 6.1 Evidence-based safety management characteristics and practices ...... 163

xiii List of Figures

Figure 1.1 Entities in the Chain of Responsibility (Source: Peter Wells) ...... 13 Figure 1.2 Chain of pressures leading to truck crashes (adapted from Williamson, 2014) ...... 16 Figure 2.1 Haddon Matrix for identifying injury factors ...... 31 Figure 2.2 Socio-technical system involved in risk management (Rasmussen, 1997, p.185) ...... 35 Figure 2.3 Haddon Matrix for Fleet Safety Management (Dubens and Murray, 2009, p. 16) ...... 37 Figure 2.4 Runyan’s 3-dimensional Haddon Matrix (Runyan, 1998, p. 304) ...... 38 Figure 2.5 Occupational Light Vehicle (OLV) Systems Model (Stuckey et al., 2007, p. 1008) ...... 40 Figure 2.6 Swiss Cheese model of accident trajectories (Reason, 1997b) ...... 41 Figure 2.7 Deviation model (Hale et al., 1997, p. 128) ...... 47 Figure 2.8 TRIPOD Beta accident causation model (Reason et al., 1989) ...... 49 Figure 6.1 Summary validation findings from Study 3 ...... 162 Figure 6.2 Model of an integrated safety management system (SMS) for heavy vehicle transport ...... 174

xiv

Extended Abstract

The aim of this thesis research was to find ways to improve safety in the heavy vehicle transport industry through the development of an evidence-based safety management system. This research was undertaken in light of disproportionate crash and injury risks associated with the heavy vehicle transport industry in comparison with other industries and other road users. No research to date has attempted to identify a set of safety management characteristics that are likely to reduce this risk. Although in recent years, work related road safety research has occurred, the problem of heavy vehicle crash related injury has largely been analysed as a public road safety road safety problem and largely dealt with through encouraging compliance to transport regulations. The nature of the trucking industry presents some unique challenges for safety management at an organisational level. This thesis argues that a “systems approach” with evidence-based safety management elements can be developed into an intervention program that is likely to improve safety outcomes in the heavy vehicle transport sector.

Drawing from the knowledge from prior workplace safety and road safety research (Study 1), a study of safety management characteristics comparing those in good safety performing heavy vehicle operators and poor safety performers sought to synthesise the distinguishing features between them. Two empirical studies were conducted (Studies 2 and 3). The first was a survey of senior managers of Australian heavy vehicle operating companies. The second was an in-depth investigation of a sample of the survey participants to validate the self-reported survey, and to learn more about the reported characteristics and non-reported characteristics in situ. The findings of these studies provided the basis upon which to build a safety management system (SMS) suitable for heavy transport vehicle operations. This process resulted in the identification of 14 safety management practices that have strong research evidence for inclusion in a safety management system (SMS) for heavy truck operations. These findings, together with analysis of sound theoretical models to underpin the SMS, were used to shape the SMS.

The SMS features three spheres of management practices – risk assessment and management, driver risk management and safety culture management. Drawing from the literature, a dynamic model of a safety management system is presented and explained. The original aim of this thesis research has been met, providing an evidence-based safety management system that is likely to reduce crash and injury risk when applied to heavy vehicle transport operations.

1 Chapter 1: Background and the Australian Trucking Industry

This thesis aims to develop a safety management system for companies that operate heavy transport vehicles by researching the characteristics that distinguish between companies with good safety outcomes and companies that have poorer safety outcomes. In this first Chapter, the specific safety problem is described together with an examination of the features of the Australian trucking industry and how these features contribute to safety outcomes.

1.1 Background and introduction

This study was initiated by road safety researchers at the University of NSW, in discussion with Australian insurers and regulators out of mutual growing concern about injuries and crash costs resulting from heavy vehicle crashes. The initial scan of the scientific literature found surprisingly little prior research into safety management practices in the heavy vehicle road transport sector. Indeed, although there had been research in the workplace safety field, there seemed scant crossover into the area of work related driving safety. Generally, approaches to safety and injury management have not been shared between workplace health and safety (WHS) and road safety sectors. For example, Knipling et al (2003) examined commercial motor carrier safety management and found that safety management system (SMS) approaches, developed and applied in workplace safety disciplines, were typically not being applied in transport management at any organisational level. This was despite the fact that WHS approaches would be most applicable to the heavy vehicle sector because it involves fleets of working drivers. Hence, an excellent opportunity existed to develop and apply SMS methodology to heavy vehicle safety in collaboration with industry, insurers and regulators. Because no study had yet built a safety management system through a research process, the outcomes from this work will advance the knowledge base in relation to road safety and workplace safety management practices and policy.

The study is also important as it addresses the major road safety problem of heavy vehicle crashes. Furthermore, whilst this study aims to strengthen truck fleet operational risk control, leading to fewer crashes, fewer and less severe injuries, reduced crash costs, greater economic efficiency and reliability in road freight transport, it ultimately aims to introduce a new way of thinking about road safety. The results of this work may lead to a new methodology of managing road transport fleets and road safety in general, and make a significant practical contribution to the wellbeing and health of the Australian population and others.

The aim of this research was to develop a new safety management system (SMS) specifically for the trucking industry. Drawing from systems theory, the work was expected to generate a paradigm shift in road safety thinking, moving beyond the compartmentalised and static epidemiological Haddon model (1968), that has underpinned road safety in many countries since the 1970s, to a new and more integrated and dynamic approach.

The novelty of the SMS approach for road safety is that it focuses on the wider context or system of road transport, recognising that hazards or risk factors are interrelated and interdependent. By focusing only on discrete contributions to crashes, the current approach in road safety fails to take into account pre-existing failures in the system that are often root causes of crashes. Taking an SMS view, transport hazards are conceptualised as a set of interacting variables that require interdependent actions to respond effectively to these risks instead of the traditional approach of a group of individual problems with sets of single interventions.

The focus of this thesis research is on the road freight sector involving transport by heavy trucks (greater than 12 tonnes in mass.)

A successful grant application to the Australian Research Council established a linkage partnership between government and industry bodies to co-finance a major study to research and develop a safety management system suited particularly for heavy vehicle transport operators. A research review team was established to ensure a high standard of research methods and data analysis. The team was led by Professor Raphael Grzebieta, Chief Investigator 1, Professor Ann Williamson, Chief Investigator 2, and Associate Professor Jake Olivier, Chief Investigator 3. The thesis author was the project manager. She was also assisted and guided by Dr Rena Friswell. The project team also reported progress on the study to a Scientific Advisory Committee that included representatives of the industry funders.

The method of the study was to identify the management characteristics that distinguish between heavy vehicle operators with better safety performance from those with less good safety performance. These characteristics then formed the basis of a safety management system on which this thesis is a major component of that study.

This Chapter provides an examination of the problem of heavy vehicle road crash and trauma risk in Australia. The features of the Australian trucking industry and regulatory environments together with the associated safety issues are discussed. Current industry and government measures taken to address safety risks in the industry are also described. This establishes the starting point in an effort to find new ways to improve safety performance in this sector.

1.2 The size of the safety problem in the road freight transport industry

The World Health Organisation estimates that 1.2 million people are killed in road crashes every year (WHO, 2015). For people aged 15-29 years, being involved in a road crash is the highest risk of being killed. Across the population, this largely preventable problem costs governments, on average, 3% of their Gross Domestic Product (GDP).

Truck crashes are a substantial part of this problem. In 2014, trucks comprised only 2.4% of the total number of vehicles registered in Australian jurisdictions and represented only 7% of total vehicle kilometres travelled, but they were involved in 19% of fatal crashes (BITRE, 2015a, b). Also, an

3 analysis of New South Wales crash data found that the crash rates for heavy trucks per kilometre travelled were not much different from the per kilometre rates for all NSW road crashes (Williamson et al., 2003), suggesting that the level of truck exposure to risk does not fully explain their overrepresentation in fatal crashes.

The figures are similar in the United States where, in 2013, heavy trucks were 4% of the registered fleet, travelled 4% of miles travelled and represented 12% of the fatal crashes and a further 4% of all injury crashes (U.S. Department of Transportation, 2015). Moreover, in Canada it is reported that in 2001 nearly 20% of road fatalities were from heavy truck crashes despite trucks only making up 4% of the registered vehicle fleet (Mayhew et al., 2004). In 2001, 524 people were killed and a further 11,574 people were injured in crashes involving heavy trucks in Canada.

Crashes involving heavy vehicles represent a workplace health and safety problem. In the 10-year period 2003 to 2012, 787 workers were killed in truck-related incidents in Australia (Safe Work Australia, 2014a). This represents 30% of all Australian worker fatalities over this period, making trucking the second highest cause of worker fatalities in Australia.

When considering truck driver fatalities alone, truck drivers were 20% of workers killed on the job over the same period, making this occupation the most fatal in Australia (Safe Work Australia, 2014b). The road freight industry generally has a notoriously high rate of worker fatalities, at 18.6 per 100,000 workers compared with the overall Australian work-related fatality rate of 1.9 deaths per 100,000 workers (Safe Work Australia, 2013).

In addition, there were 4,000 non-fatal workers compensation claims per year from 2002-2011 in the Australian road freight industry (Safe Work Australia, 2013). These serious injuries are those involving a permanent or temporary incapacity that require a week or more off work. In the year 2002- 2003 the rate of serious injuries in heavy vehicle transport was 36.8 per 1,000 workers, and 29.8 in 2009-2010. The median time lost, as well as cost per workers’ compensation claims resulting from truck drivers’ injuries, are consistently higher than the figures for all other industries.

In the US, an examination of work related deaths and injuries found that truck driving accounted for 12 per cent of all worker deaths, accounting for more fatalities than any other occupation. Truck drivers also accounted for more non-fatal injuries than all occupations (Knestaut, 1997). Similarly, Canadian government figures for work-related fatalities of workers in federal jurisdiction employers found that in 2011 around 60% of these fatalities occurred in the road transport sector (Employment and Social Development Canada, 2014). Furthermore, it was reported that 6,556 road transport workers sustained disabling injuries on the job that year.

The total cost to the community is also substantial. The cost of truck crashes has been estimated to be around $37 billion per annum in the US (Lueck, 2011) and around $3.8 billion per year in Australia (Centre for International Economics, 2011).

1.3 The dimensions of the problem

A great deal of research has been undertaken to determine the main contributing factors in heavy vehicle road crashes and associated injury. Examination of data collected for 967 heavy vehicle crashes in the US (FMCSA, 2006) found that driver factors included the following: driver inattention, fatigue, drug impairment, decision errors, and speeding. Driver factors were determined to be critical reasons for the crashes in 87% of cases. It also found that vehicle factors (such as brake problems) were important, and were present in 10% of cases. These were mostly related to brake, tyre and wheel problems. Environmental conditions, such as road, traffic and weather, were critical reasons in 2% of cases.

Research on specific risk factors in truck crashes – especially driver fatigue (Feyer et al., 2002), speeding (Brooks, 2002), driver pay (Belzer et al., 2002), and vehicle factors (Blower et al., 2010) – has helped to focus attention to some of the more prevalent risk factors. Bezwada (2010) found that the risk of driver fatigue, drowsiness, and inattention were more predominant in truck drivers than in other motor vehicle drivers. Moreover in a study of fatal truck crashes in the State of Victoria between 1999 and 2007, the drivers in one in six of these crashes were found to have stimulant or other drugs in their systems (Brodie et al., 2009). This study also found that one third of these fatal crashes involved a single vehicle leaving the roadway on a straight stretch of road, and that nearly one quarter of the crashes involved excessive or inappropriate speeds for the conditions. A NSW study also found that speed and fatigue were prominent factors in truck crashes (Boufous and Williamson, 2006).

It is well established that driver fatigue is one of the most prevalent factors in truck crashes (Crum and Morrow, 2002; Dingus et al., 2006; Feyer and Williamson, 1995; Feyer et al., 2002; Hanowski et al., 2009). Data from a large study of 549 insurance reports of major truck crash investigations (defined as crashes with losses costing $50,000AUD or more) in Australia found that speeding was involved in 27% of the crashes, with fatigue the next greatest contributor to these crashes at nearly 13% (NTARC, 2015).

Some studies have tried to determine some of the underlying conditions in which these risk factors manifest. For example, Richards (2004) found that motivators for truck drivers to use drugs included fatigue, peer pressure, wanting to fit the trucking “image”, socialisation, relaxation and addiction. Also, Kemp et al (2013) found that time pressures can lead to physical fatigue and emotional exhaustion, which in turn lead to negative attitudes about compliance with hours of service regulations. These kinds of analysis help us to understand why crashes occur and go deeper towards finding the root causes of serious truck crashes – but perhaps not deep enough. The “why” question needs to be extended even further. Why, for example, do drivers feel excessively time-pressured? Why do truck drivers need to take drugs?

5 1.4 Road freight transport industry structure and inherent pressures on safety

This section examines the nature of the job of truck driving, the nature of the road freight industry and the systemic injury risk factors that underlie the manifestations of crash and injury causation.

1.4.1 Basis of risk in the truck driving occupation Truck driving is a demanding occupation with long and irregular shifts and work pressure resulting in high driver fatigue risk and mental stress (Friswell and Williamson, 2010). American researcher, Professor Michael Belzer (2000), challenged his readers to “imagine a world” where there is no minimum wage, most work 60 hours per week on average, most workers have to compete to offer services at the lowest possible price, the work involves irregular shifts and hours – both day and night – and employers decide what to pay for and what tasks must be performed for no pay. This he says is the real world for most truck drivers. This description is supported by studies in Australia as well. For example, Mayhew and Quinlan (2006), in interviews with 300 long haul Australian truck drivers, found that economic pressure together with the expansion of contingent work arrangements in the trucking industry has had a negative impact on WHS outcomes.

1.4.1.1 Risks associated with truck driver pay methods Whether working as an employee driver or a contractor, truck drivers, like other workers, try to optimise their financial benefit through choices they make about their work practices. The method of driver payment influences the extent to which drivers engage in elevated risk behaviours. An American study found that unregulated hours of work and unpaid non-driving work (such as loading, unloading or waiting for loads) provide incentives for drivers to work longer hours and to undertake risk fatigued driving (Arboleda et al., 2003).

The method of payment has been linked to truck driver behaviour and to safety outcomes. Drivers can either be paid for the hours that they work, or they can be paid on a productivity basis. The “productivity” method of payment is a compensation method that ties financial compensation to output, either by truckloads delivered, kilometres driven, or profits earned by a job. Under this type of payment method, the employer may or may not pay for time spent on non-driving activities such as loading, unloading or queuing/waiting. Sometimes a flat fee is given to the driver for some or all of these tasks. Sometimes payment for time the driver spent waiting is conditional on the duration of time the driver spends waiting, e.g. drivers get paid for time after the first hour. The payment for time worked and productivity payment methods can be combined in other ways as well, such as having drivers on hourly pay and also providing them bonuses as a share of the profits earned by a company.

The way in which drivers are remunerated influences the likelihood of unsafe behaviours and crashes. Table 1.1 describes six studies from the 1990s to 2014 that provide evidence of how pay methods affect drivers’:

• self-imposed time pressure;

• use of stimulant drugs;

• speeding;

• fatigue; and

• truck maintenance and safety checks.

Productivity-based pay is found to produce incentives to self-impose time pressure, take stimulants, speed and work excessive hours. Productivity pay also predicts driver fatigue and encourages drivers to risk fatigued driving, poorly maintain trucks and skip safety checks.

In a survey of 573 US motor carrier drivers in 1997, Monaco and Williams (2000) found that hourly payment for drivers had a 10.2% lower crash risk compared with productivity pay, i.e. when drivers are paid by the mile or as percentage of revenue earned by the company. Moreover, where drivers are paid mileage rates, a $0.10 increase in the rate results in a 1.76% reduction in the risk of crashing.

Additionally, a study by Williamson and Friswell (2013) found that nearly 90% of Australian truck drivers have wait to load or unload their trucks, but just one quarter of them are paid while waiting. They found that these drivers experienced more fatigue than did drivers who were paid to wait. The researchers concluded, “mandating payment of drivers for non-driving work including waiting would reduce the amount of non driving work required for drivers and reduce weekly hours of work. In turn this would reduce driver fatigue and safety risk as well as enhancing the efficiency of the long distance road transport industry.” As driver fatigue and speeding are the major behavioural contributors to truck crashes, the influence of truck driver pay methods is an important one to examine further.

7 Table 1.1 Effects of Driver Payment Methods on Risk Behaviour Study focus Author, year Method/sample Findings Effects of driver pay system (Golob and Cross-sectional Drivers try to optimise money earned by on propensity to speed, self- Hensher, 1994) survey/ n = 402 self-imposed time pressure, leading to impose tight schedules, take Australian truck use of stimulants, leading to speeding. stimulant drugs drivers (79% are paid based on productivity) Effects of driver pay method (Hensher and Cross-sectional Non-drug users drive 20 km/h slower on propensity to speed Battellino, 1990) (pilot) survey/ n than drug users. = 46 Australian Drivers paid on a percentage of truck truck drivers earnings drive 15 km/h faster. Effects of productivity based (Williamson et Cross-sectional Drivers paid by amount of work done payment on driver fatigue al., 2001) survey/ n = report fatigue more often than drivers 1,007 Australian paid by the amount of time they worked. long haul truck drivers Effects of compensation (Arboleda et al., Cross-sectional Unregulated hours of work and unpaid methods on driver fatigue risk 2003) survey/ n = 116 non-driving work provide incentives for US trucking drivers to work longer hours and risk companies driver fatigue. Effects of payment methods (Williamson, Re-analysis of 2 Drivers paid by productivity were 2-3 on drug use 2007) Australian times more likely to use stimulant drugs. surveys 7 years apart/ n=970 & n=1007 Effects of payment methods (Thompson and Cross-sectional Performance based pay encourages on driver fatigue Stevenson, survey/ n = 346 drivers to keep driving at the expense of 2014) Australian truck sleep and rest, maintenance and safety drivers checks. Effects of payment methods (Williamson and Cross-sectional Incentive based payment and unpaid and unpaid tasks on driver Friswell, 2013) survey/ n = 475 waiting times predict driver fatigue. fatigue Australian truck drivers

1.4.1.2 Risks associated with truck driver pay levels There is also evidence that the level of payment for truck drivers, i.e. how much money truck drivers are paid, influences their behaviour and health/safety outcomes. Three cross-sectional studies and two cohort studies carried out in 3 countries from the 1980s to 2014 (Table 1.2) provide evidence that pay levels affect drivers’:

driver turnover; speeding; self-imposed schedules; stimulant drug use; and violations of work hour limits.

One large cohort study (Belzer et al, 2002) found increases in pay reduced crash risks. Moreover, another study (Williamson and Friswell, 2013) found that unpaid waiting time and incentive based pay predict driver fatigue. Also, in the early 1980s, Scandinavian researchers found strong links between driver pay and driver turnover (Backman and Järvinen, 1983). The most common reasons that drivers gave for changing their work were: unsatisfactory salary (31%), heaviness of the work (20%, irregularity of working hours (14%) and health (12%).

A survey of 820 Australian truck drivers, carried out in 1990, concluded that economic rewards were a major influence on drivers’ decisions to speed on delivery journeys (Hensher et al., 1991). Work practices of truck drivers, including speeding, self-imposing tight schedules and taking stimulant drugs are encouraged by uncertain or insufficient earnings. These findings were later replicated in two further surveys of Australian truck drivers, in 1991 (n = 970) and in 1998 (n = 1,007), both confirming the influence of productivity payment systems on the use of stimulant drug use by drivers (Williamson, 2007). Golob and Hensher (1994) highlighted the concern that inadequate understanding of the relationship between trucking industry characteristics and on-road safety performance may lead to inappropriate and ineffective regulatory responses. Their study involving interviews with 402 Australian truck drivers examined the constellation of endemic pressures on drivers to speed on delivery journeys. They concluded that rates of financial rewards influence drivers’ propensity to speed, self-impose unsafe schedules and take stay-awake pills. They observed a complex relationship of decisions by drivers to optimise financial gains through a number of factors that impose difficult timeframes on themselves, which in turn encourages the use of stimulant drugs, which in turn is associated with speeding. The majority of drivers (79%) were paid directly in relation to the earnings of the truck.

In parallel, major surveys of US truck drivers found links between driver pay and safety performance. Braver et al (1992) found that two thirds of drivers report that they violate weekly hours of service restrictions and that low pay rates and tight delivery schedules were a major impetus to this violation. Moreover, Belzer et al (2002) concluded that drivers who are paid a higher rate have significantly fewer crashes after estimating from the data that a 10% increase in driver pay from $0.295USD per mile to $0.324 per mile reduced the probability of a crash by 21% from a 13.8% chance to a 10.86% chance.

As shown in Table 1.2, studies in three countries have shown correlations between pay levels for truck drivers and their behaviour, health, safety and crash risks.

9 Table 1.2. Effects of Driver Pay Levels on Safety Behaviour and Outcomes Study focus Author, year Method/sample Findings Effects of driver pay levels on (Backman and Cohort study/ n = Reasons for leaving job: driver turnover and health Järvinen, 1983) 472 drivers in 31% unsatisfactory salary, Finnish Transport 20% work too heavy, Workers Union 14% irregular hours, and 12% health affected. Effects of driver pay levels on (Hensher and Cross-sectional Drivers paid on a percentage of propensity to speed Battellino, 1990) (pilot) survey/ n = truck earnings drive 15 km/h faster. 46) Australian truck drivers Effects of driver pay levels on (Hensher et al., Cross-sectional Freight rates for owner-drivers propensity to speed, self- 1991) survey/ n = 820 influence speed. impose tight schedules, take Australian truck Uncertainty of income encourages stimulant drugs drivers self-imposed schedules and drug use. Effects of driver pay levels on (Braver et al., Cross-sectional Low pay and tight schedules violations of work hours 1992) survey/n = 1,249 US predict violations of work hour truck drivers limits. Effects of driver pay increases (Belzer et al., Cohort study/ n = A 10% increase in pay reduced on crash involvement 2002) 11,540 drivers crash risk by 21% A 10% increase (Rodríguez et employed by J.B. in paid days off reduced crash risk al., 2003) Hunt (US) by 7%. (Rodriguez et For every additional cent per mile al., 2006) paid to a driver, the crash count decreases by 8%. A 1% increase in pay corresponds to a 1.33% reduction1 in crash risk probability.

Regarding health risks, according to a study of 1,047 Australian truck drivers, sleep disorders, smoking and obesity are health problems more prevalent in truck drivers than in the general population (Elkington and Stevenson, 2013). Furthermore, a survey of 300 Australian long distance truck drivers found that occupational violence was endemic in the trucking industry, including verbal abuse and road violence from other motorists, abuse and threats from staff at freight depots, and abuse from customers (Mayhew and Quinlan, 2001). Long distance owner-drivers reported that they were more often subjected to this abuse than employee drivers. This survey of drivers also highlighted long working hours, endemic fatigue, and a heavy burden of injury, ill health and stress among this work group, with owner-drivers carrying the brunt of this burden. These problems were attributed to economic pressures and a lack of regulation in supply chains to protect drivers’ workplace health and safety (Mayhew and Quinlan, 2006). A Dutch study found that occupational stress and fatigue in truck drivers were related to high workload demands and a lack of job control by drivers (De Croon et al., 2002).

Light truck and short haul drivers are not exempt from these economic pressures and adverse health and safety consequences. Indeed, a study of light and short haul drivers found that stress and fatigue

1 The safety benefit of increased pay levels does not reduce over time, but the effect reduces incrementally as rates of pay become higher.

and associated injuries and illnesses are also common in this sector (Friswell, 2013). Moreover, work organisation, road and site access, vehicle/equipment ergonomics, and interpersonal conflicts are associated with injury and health risks for these workers (Friswell and Williamson, 2010)

Based on available workplace health and safety data, truck driving is the most lethal occupation. Between 2002 and 2012, on average, Australian truck drivers were 20% of work related fatalities, with farm workers the second highest percentage at 18% of work related fatalities (Safe Work Australia, 2014c). This situation appears unchanged in recent decades, despite increasingly tighter road transport regulation. The truck driver remuneration methods and levels are an important influence on safety. Some of the underlying reasons for the continuing high levels of trauma in the industry may be found in the relative economic and social powerlessness of drivers. The next section describes the industry context of this problem.

1.4.2 Road freight transport industry structure and inherent safety risks In order to provide some context for the inherent occupational risks for truck drivers it is important to examine the road freight industry itself. A major inquiry into safety in the Australian trucking industry concluded that “…commercial/industrial practices affecting road transport play a direct and significant role in fomenting hazardous practices” (Quinlan, 2001, p. 22). The Quinlan report found evidence of a strong association between commercial practices and safety, as well as industry incentives to breach safety regulations. Tendering practices in a competitive environment have pushed freight rates down to nonviable levels and encouraged small companies and drivers to engage in dangerous practices, such as speeding, working excessive hours, skipping truck maintenance, overloading and using stimulant drugs (Williamson, 2007).

Similarly in the US, when motor carrier charge out rates were deregulated in the 1970s, trucking companies became less profitable, unionisation plummeted and driver remuneration dropped, resulting in a lower standard of safety (Belman and Monaco, 2001).

In 2006, the Australian National Transport Commission (NTC) engaged a business consultant to review opportunities to improve efficiencies in the Australian road freight industry in the light of forecasts of a doubling of the freight task (measured in tonnes per kilometre travelled) in the period 2000-2020 (Manders, 2006). The report was sceptical about the possibilities of improving safety practices with continuing economic and efficiency pressures facing the industry. The consultant found similar economic and efficiency pressure problems in Europe and North America.

Australia’s freight volumes have increased six-fold over the past 40 years (BITRE, 2011). Road transport accounted for 8.6% of Australia’s GDP in 2013. Industry representatives predict that the freight task is expected to increase by 80% between 2010 and 2030 (Australian Logistics Council,

11 2015)2. There have been productivity gains through the use of larger vehicles such as B-double articulated trucks, but there is a concern that future productivity gains may not be sufficient to cope with growing demands on the industry.

The Australian Logistics Council (2015) predicts that total demand in the industry is expected to increase, and revenue is projected to grow at a annual rate of 2.9% over the next five years. However, industry profitability is expected to decline as operators continue to face skills shortages and potentially higher access costs to the country's road networks, while fuel prices are expected to rise from current low levels and increase cost pressures. Profit margins in the industry are already generally low and competition is fierce (IBISWorld; Mayhew and Quinlan, 2006; Quinlan and Wright, 2008a): this problem is forcing many smaller operators to sacrifice profit margins or leave the industry. Industry bodies such as the Australian Logistics Council (Australian Logistics Council, 2015) and the Australian Trucking Association (Price Waterhouse Coopers, 2013) report that the growth of the freight task and customer demands for meeting tight timelines are prompting the use of more technology. In addition, there is increasing competition with other industries for recruiting skilled drivers.

The road freight transport industry in Australia comprises mainly small transport companies with fewer than 20 trucks. In fact, more than 99% of companies have fewer than 20 trucks, and around 70% are single truck owner-operators. Less than 0.5% of trucking companies have fleets of more than 100 trucks (ACIL_Tasman, 2004). The industry has complex layers of subcontracting, remote work without direction supervision, and irregular and shift work (Mayhew and Quinlan, 2006). This may be due in part to the dominance of the small number of large companies. With 14% of the transport market serviced by three companies (Ferrier Hodgeson, 2014), these companies would employ a large proportion of the Australian truck driving workforce.

There is also a concentration of road freight transport buying power. Over 32% of revenues generated in the Australian transport sector are from retail and wholesale road transport clients with two retail companies commanding 70% of this market (Australian Industry Group, 2012). In a determination of the NSW Industrial Relations Commission, the problem with this scenario is described citing evidence from the Australian Transport Workers Union as follows.

At the top of the contractual chain is the Head Consignor - large companies with very substantial transport requirements - such as Woolworths and Coles. Traditionally, these entities do not undertake transport work directly, but contract out their requirements on a cost competitive basis. The next tier of the chain is the Consignor, the major transport companies such as Toll Transport,

2 The Australian Logistics Council is an industry association founded by the largest retail and transport companies to lobby for improved regulatory conditions. ALC focuses its advocacy efforts on five key areas with the aim of improving supply chain efficiency including: supply chain logistics safety, infrastructure, regulation, technology, & people.

Patricks, Linfox and TNT. These entities may utilise their own direct labour or subcontract out the work to the third tier in the chain - the smaller transport companies. These companies may then contract out the work to individual contract drivers or carriers.

The Union submitted that while the main problem areas for safety are to be found at the third and fourth levels of the chain, the competitive pressures exerted by the Head Consignors and Consignors meant that these entities must bear some responsibility for the problems which exist down the line (Wright and Walton, 2006, p.10)

This case, for a new award and contract determination, was an application by the Union to require safe driving plans, agreed by all parties in the transport chain, as well as other requirements, including induction training, and drug and alcohol policies. Australian transport law assigns responsibility to each entity or person in the chain of work involved in delivering goods where their actions or inactions can influence safety in the process. The decision in this case recognised that the pressures exerted by the Head Consignors and Consignors were in part responsible for the problems that manifest further down the chain and that there was a need for a specific legal instrument to hold these parties to account. The roles in the road freight transport and logistics chain is depicted in Figure 1.1

Figure 1.1 Entities in the Chain of Responsibility (Source: Peter Wells)3

The chain of responsibility provisions are now embedded in the Australian Heavy Vehicle National Law (HVNL). The aim of these provisions is to assign responsibility to all duty holders in the transport and logistics chain for the safety of the whole delivery process. Duty holders must ensure that “terms of consignment or work/employment contracts will not result in, encourage, reward or provide an incentive for the driver or other party in the supply chain (e.g. a scheduler) to break the HVNL4”.

3 NSW Roads and Maritime Services http://www.slideshare.net/informaoz/peter-wells-42284453 accessed 6 March, 2016) 4 For more information see https://www.nhvr.gov.au/safety-accreditation-compliance/chain-of-responsibility/about-the-chain-of- responsibility

13 The Union’s case for safe driving plans was partially about the disproportional influence of large transport customers to set unsafe freight rates, thereby influencing the unsafe behaviours discussed in Section 1.4.1 above. The argument is that market forces alone could not address these safety risks, when drivers and small transport companies are pressured to accept low rates of pay. Therefore additional regulatory safeguards were needed.

Outsourcing and subcontracting are very common practices in the road freight transport industry. In part, this may be because transport contract work is variable in volume requiring more flexibility in hiring practices. Whether or not transport companies or consignors deliberately attempt to diffuse their responsibilities for managing safety through subcontracting work, the WHS effects of outsourcing (such as changes to injury rates and severity, WHS knowledge and compliance) are poor WHS outcomes (Johnstone et al., 2001; Quinlan et al., 2001).

A major inquiry into safety in the Australian trucking industry took submissions from a large number of industry experts and researchers who consistently advised that intense competition, industry tendering practices, low freight rates and pressure from clients were probably the most fundamental source of dangerous practices in the industry (Quinlan, 2001). Part of the Inquiry entailed a survey of drivers (n = 300). The results indicated that there is a range of physical and psychological health afflictions as well as low-level occupational violence that are disproportionately reported by truck drivers with more prevalence and severity, affecting owner-drivers.

There was evidence to the Quinlan Inquiry, in a submission from an insurer, that many operators were not financially viable; and in fact the commercial environment for the industry was such that it put into question the financial viability of the industry as a whole. Downward pressures on freight rates meant that the rates were so low as to pressure drivers and companies to push the margins (meaning less truck maintenance, more trips, longer hours, speeding, etc.) In summary, it was found that the existing mix of transport and workplace safety regulatory authorities, and laws and regulations were found to be less than effective in enforcing safety regulations in this industry compared with other industries. The Inquiry concluded that coordination and resourcing of regulatory activities in relation to safety in the long distance trucking industry are major issues that should be addressed as a matter of urgency.

Indeed, the Quinlan Report of Inquiry into the Long Distance Trucking Industry concluded that:

Tendering practices common in the industry contained a number of elements clearly not conducive to safe operation. For example, tenders often took little explicit account of how a task was to be completed or other safety related issues and often quoted ‘all in’ prices that placed cost burdens on the transport company even for events beyond its control or due to customer inefficiency. Contracts often did not impose/enforce waiting time charges meaning that the customer had no incentive, other than their own convenience, for unloading trucks promptly.

Given that local delivery drivers were paid on an hourly basis there was often an incentive to leave long distance trucks waiting. Delays exacerbated pressure to arrive early to beat the queue or race to get to the next job, especially amongst owner/drivers but also fleet drivers (Quinlan, 2001, p.21).

In summary, the Inquiry concluded that coordination and resourcing of regulatory activities for safety in the long distance trucking industry were major issues that needed to be addressed as a matter of urgency.

Mayhew and Quinlan (2006) further examined the precarious nature of employment in the trucking industry and found that this was detrimental to drivers’ workplace health and safety. Furthermore, owner-drivers – who make up the majority of Australian trucking companies – were found to have worse WHS outcomes than employee drivers. And while owner-drivers make up more than 60% of operating entities in the industry, they account for only 11% of the income earned in the industry (ACIL_Tasman, 2003). Increasingly Australian jurisdictions have applied chain of responsibility principles to transport regulations and, to a lesser degree, workplace safety legislation. This has been claimed as a step in the right direction for enhancing safety in the transport industry (James et al., 2007).

Industry pressures in the Australian transport industry are strong. In a hearing of the NSW Industrial Relations Commission which determined the Mutual Responsibility for Road Safety (State) Award 2006 (Wright et al., 2006), the introductory statement by the presiding judge quoted another judge at length. An excerpt is quoted here. His Honour Judge Graham commented in the District Court in August, 2005, when sentencing a long distance truck driver to jail for causing the deaths of two road users in a crash:

Timetables and deadlines which make inadequate provision for rest for drivers, particularly if the deadline is at risk of being missed, and timetables which expect drivers to maintain a pattern of driving overnight from one capital city to another, six nights a week [sic]. Over the years, various measures have been implemented with a view to seeking to restrict the effects of such business practices.

In the present matter, the statement of facts refers to safety cams 5 and logbooks. Restrictions on the maximum speed of heavy vehicles have also been implemented. Despite those measures, heavy vehicle truck drivers are still placed under what is, clearly, intolerable pressure in order to get produce to the markets or goods to their destination within a time fixed, not by any rational consideration of the risks involved in too tight a timetable, but by the dictates of the marketplace (Wright et al., 2006, p.6).

5 “Safety cams” are devices that are used by regulatory authorities to measure time and distance of travel by individual trucks and their drivers.

15 The determination that resulted from this hearing required the adoption of safety management responsibilities for companies contracting road transport services, including a provision requiring transport operators to prepare a safe driving plan for their subcontract drivers, and for consignors to enter contracts that agree to strict compliance with the safe driving plans (Wright et al., 2006).

In summary, the picture of trucking safety is not a very positive one. The Australian trucking industry is currently characterised by inherent safety risks. Figure 1.2 depicts a model of the trucking industry pressures that link risk and crash outcomes supported by the findings of a recent literature review (Mooren et al., 2015).

Industry Company Work Risk Outcomes pressures responses Conditions behaviors

Figure 1.2 Chain of pressures leading to truck crashes (adapted from Williamson, 2014)

This model proposes relationships between factors that interact to increase the risk of crashes in the heavy vehicle transport industry. The road freight transport industry exhibits characteristics that lead to transport company actions that in turn lead to unsafe working conditions. These industry features and actions result in unsafe driver behaviours and ultimately lead to adverse safety outcomes. Key elements of industry pressure, company responses, working conditions and risk factors are summarised below.

1.4.2.1/ndustry Pressures At a macroeconomic level the industry pressures include intense competition among trucking operators, tight profit margins and long chains of contracting and sub-contracting (Mayhew and Quinlan, 2006). It can be argued that competition forces downward pressure on profit margins that in tum lead to unsafe practices. Their survey of 300 Australian truck drivers conduded that this ultimately results in poorer outcomes for drivers in smaller companies than for employee drivers larger companies. Comparing rates

*Keynote presentation at Occupational Safety in Transport Conference, Gold Coast, 2014.

of involvement in major road crashes, 10% of owner drivers and 13% of small fleet drivers reported involvement in major crashes compared with 6% of large fleet drivers (Mayhew and Quinlan, 2006).

1.4.2.2 Company Responses Company responses to low profit margins are often to pay drivers based on productivity. In a survey of heavy truck drivers travelling through New South Wales, more than three quarters of drivers reported being paid by this method in Australia (Williamson and Friswell, 2013). Research has shown relationships between productivity-based pay and safety-related outcomes in this industry, with evidence that productivity pay encourages drivers to work more hours or take on more jobs, leading to the use of stimulant drugs to combat fatigue (Williamson, 2007). In addition, it is argued that low profitability encourages companies to keep trucks operating when the work is available even when this means skipping maintenance, resulting in defects and unsafe trucks on the road.

In addition, for companies to achieve the operating flexibility to maintain profitability, they very often subcontract drivers to keep their wages commitment as low as possible. With intense competition, trucking companies and drivers are under pressure to accept unrealistic contracts in terms of price and agreed delivery timeframes, which in turn impose time pressures on long journeys, often resulting in driving at higher speeds and driving while too tired (Quinlan and Mayhew, 2001).

1.4.2.3 Work Conditions All of the industry pressures, and company responses to these pressures described in Figure 1.2, provide an environment where drivers are under time pressures, drive long distances, work more hours, in vehicles that have mechanical defects (Hensher et al., 1992). Consequently, there is high driver turnover and job dissatisfaction in the industry (De Croon et al., 2002; Hensher and Battellino, 1990; Min and Lambert, 2002).

1.4.2.4 Risk Behaviours and Outcomes In the scenario described above errors and violations are likely to occur. Speeding, fatigue, use of stimulant drugs, vehicle defects, unsafe trucks are all major crash and injury risk factors (Golob and Hensher, 1994).

A survey of 715 union members (Transport Workers Union of Australia, 2012) found evidence of significant economic pressures on drivers and risky practices. The survey results showed that:

48 per cent of drivers reported almost one day a week in unpaid waiting time. For delivery drivers, it was more than 10 hours a week; 56 per cent of owner-drivers had to forgo vehicle maintenance because of economic pressure, the need to keep working or the high cost of repairs; 27 per cent felt they had to drive too fast, and nearly 40% felt pressured to drive longer than legally allowed; many saying that the pressure [to speed and drive excessive hours] came directly or indirectly from the client.

17 The disproportionate levels of fatigue in truck drivers are well documented and explained by irregular and unhealthy sleep and rest patterns experienced by drivers due to schedules and general work environments that are too often not conducive to restorative sleep (Adams-Guppy and Guppy, 2003; McCartt et al., 2000; Moreno et al., 2004; Williamson et al., 2001). And while there is little evidence of disproportionate use of alcohol by drivers, the effects of even moderately fatigued driving in terms of decrements to performance have been found to be equivalent to illegal and unsafe levels of intoxication (Williamson and Feyer, 2000). In addition, there are elevated morbidity patterns including obesity, diabetes, cardiovascular disease, sleep disorders, cancers, musculoskeletal disorders, arthritis, chronic back pain, and depression – all related to the environmental conditions that characterise the trucking industry (Apostolopoulos et al., 2012).

Increasingly transport safety research is identifying the detrimental effects of systemic pressures, such as contingent work arrangements, low job security and low pay, on truck driver health and safety (Mayhew and Quinlan, 2006). With regard to the contingency-work effects on short haul drivers, Williamson et al (2009) found that main predictor of both illness and injury for all short haul truck drivers was work-life conflict.

In summary, the road freight industry is fraught with safety vulnerabilities owing to the nature and structure of this industry.

1.5 Safety regulation of the road freight transport industry

The heavy vehicle road freight industry is subject to transport safety regulations and workplace health and safety (WHS) regulations in Australia. The aim of transport safety regulations is to protect public safety. That is, in the public road environment, road and transport authorities regulate the use of the road to protect people from being harmed on public roads. In the case of transport regulations these initially focused on regulating individual road user behaviour through a licensing system with operating requirements and restrictions, as well as regulating access to motor vehicles through regulatory restrictions on vehicle types, conditions and features.

In the 1990s, regulating agencies began to recognise that, in the heavy vehicle transport industry there were important safety influences on truck drivers that were not being addressed with existing transport regulations. The National Road Transport Commission (now NTC) was established to reform transport regulations and to harmonise these across State jurisdictions. By 2001 a reform Bill was introduced to include chain of responsibility principles (McIntyre and Moore, 2002). This proposed principle has been incorporated into the HVNL. It enables enforcement to go beyond drivers, and places responsibilities on others in the supply chain, including employers and customers.

A comparison of the transport regulatory systems in the United States (US) and Australia (Mooren et al., 2012) was carried out in light of similar heavy vehicle safety performance for those jurisdictions, despite the much better general road safety performance in Australia compared with the USA. The

heavy vehicle transport regulation frameworks are similar in Australia and the US in some respects, but there are some key differences. Australian regulatory bodies pose restrictions relating to the types, conditions and features of vehicles permitted to travel on Australian public roads. There are also specific licensing restrictions applying to drivers of heavy vehicles, such as hours of service, speed regulations, and other road rules. Similar restrictions are in place in the US. The main difference is that in the US, companies that operate heavy vehicle transport are separately registered to do so and are audited to ensure that minimal safety management practices are in place.

In Australia, there is no system of registering heavy vehicle transport companies. Practically anyone can set up a transport company. The regulation of the industry in Australia is focused on vehicle standards, mass and load restrictions, road access and driver licensing, with professional drivers needing to meet standards for various classes of vehicles that they drive and various types of product they deliver.

The US system is more prescriptive and direct in what they require of heavy truck operators than the Australian counterpart. In Australia, companies can be held accountable for deficiencies in safety management practices, but for the most part, enforcement of their responsibilities is usually after some serious vehicle or driver regulatory breach is detected. This is probably because safety management practices are neither clearly nor specifically defined in regulations. Moreover, the safety performances of companies and drivers is not made known to the public in Australia as it is in the USA.

The Compliance and Enforcement Bill prepared by the National Transport Commission in 2003 (NTC, 2004) introduced a legal instrument to prosecute any of those involved in the transport and logistics chain who influence breaches of the transport laws. This is termed the “chain of responsibility” principle. This instrument, which was enacted, has now been adopted in Australian State legislation, giving authorities the ability to investigate company practices and individuals’ behaviours and prosecute any entity in the transport and logistics chain for safety breaches. However, specific management responsibilities are not defined in the legislation. For example, the National Heavy Vehicle Regulator (NHVR) advises that all parties in the transport chain must “make sure the terms of consignment or work/employment contracts will not result in, encourage, reward or provide an incentive for the driver or other party in the supply chain (e.g. a scheduler) to break the HVNL.” But the Heavy Vehicle National Law (HVNL) does not say what specific actions or inactions the parties should undertake, or not undertake in order to comply, other than saying that contracts that require drivers to break the law are illegal.

There has been a sustained effort to harmonise road transport legislation, and in most Australian States, Police and road authorities work collaboratively in enforcement operations. However, the Victoria Police advise that the chain of responsibility provisions in transport law are too weak, citing that the inclusion of phrases like “so far as is reasonably practicable” allows “for the crooks to

19 establish a defence,” particularly the larger companies (Skinner, 2015). They also argue that heavy vehicle legislation is overly complicated and changes are made so often that the lack of consistency in law can be used as a defence, while smaller operators are ill placed even to understand the laws. In a submission to the NTC, Victoria Police say “it’s hard for authorities to act before an accident actually happens; penalties imposed by courts are generally only 2-5 per cent of the maximums; and compliance with an industry code of practice should not of itself be a reasonable steps defence.”

In effect it is easier to enforce measures against individual driver breaches and vehicle defects than remedy poor organisational and industry practices. The Chain of Responsibility provisions enable investigations throughout transport and logistics chains, but they do not specify particular responsibility requirements. So far, very few customers have been prosecuted under these provisions, limiting their effectiveness. Perhaps this is due to the difficulties of enforcing them. The Victoria Police admit that this is a challenge they face (ATN, 2015). Also, while Australian transport legislation is starting to proscribe safety management responsibilities, there are no defining actions or systems defined in laws. A recent review of the Chain of Responsibility provisions and public discussion paper highlighted the need to address these issues (NTC, 2015).

In addition to the existing transport laws, the Australian Government passed the Road Safety Remuneration Act in 2012 under industrial relations and WHS regulatory framework. This Act gave powers to an independent Tribunal established to determine safe rates for the Australian trucking industry (Acton, 2012). The Tribunal had the powers to make remuneration orders, assist with collective bargaining agreements, resolve disputes and conduct research into pay, conditions and other matters related to trucking safety and remuneration. This meant that trucking companies could potentially be told they must pay specific rates to drivers. Indeed, the last order made by the Tribunal was to establish minimum rates for subcontracted drivers. However, on 19 April, 2016 in a special sitting of Parliament, the Government abolished the Road Safety Remunerations Tribunal following protests by some sections of the industry that their businesses would suffer if their clients had to increase rates paid to them.

Workplace health and safety regulation begins with a focus on employers’ responsibility for providing a safe work environment. The work environment can include a motor vehicle, but these regulations do not necessarily identify trucks and road specifically as workplaces for truck drivers. However, companies that employ drivers are subject to safety regulations while the drivers are working in those environments. Specifically, all employers including transport companies must, under Australian WHS Acts, show that they are diligent in ensuring health and safety of their workers (Comcare, 2011). Under Section 27 of the WHS Act, they have to be able to show that they have taken reasonable steps in relation to:

keeping up to date on current knowledge of health and safety matters,

understanding the safety hazards and risks associated with their operations, ensuring that appropriate resources and processes are applied to minimise risks, ensuring prompt responses to incidents, hazards and risks, and having and using processes for ensuring compliance with duties or obligations under the WHS Act.

In Australia, while heavy vehicle operators are subject to workplace health and safety regulation as are employers in all industries, in practice WHS authorities tend to focus their priorities on the risks at worksites, such as manual handling of freight, falls, securing loads, onsite traffic management, wellness, and returning to work after an injury as indicated in a government publication on key safety issues for employers to address in the road freight transport industry (WorkCover NSW, 2015). Moreover, there is no specific mention of driving as a work-related task, nor anything about road injury risk, in the Model Work Health and Safety Regulations6 despite truck driving being the most fatal occupation, although drivers and their employers are covered by these regulations.

1.6 Compliance and alternative compliance schemes

Sometimes the challenge for governments is to manage competing objectives. The road transport industry needs to be efficient to meet the demands of a growing freight task while also meeting the needs of improved safety performance.

In Australia, in the early 1990s, there was a growing brief that accreditation could be an effective means of demonstrating compliance with road transport law, as well as maintaining or improving efficiencies in road freight transport. The National Road Transport Commission released discussion papers on the importance of appropriate privileges-based and incentives-based strategies, including safety management accreditation, in 1994 and 1998 to complement the sanctions-based models of compliance (McIntyre and Moore, 2002).

An example of this move towards alternative approaches to compliance was the development of new approaches to fatigue management. This led to a Queensland experiment with a Fatigue Management Program introduced in 1996. The pilot project was based on the objective of adopting ‘performance-based’ safety management, rather than rule-based approaches. Those heavy vehicle transport operators who sought to be accredited under the alternative compliance “Fatigue Management Program” pilot were required to demonstrate that their scheduling and rostering processes took into consideration: each driver’s previous working time, schedule and roster; safe driving time and work activities; vehicle suitability and roadworthiness;

6 See http://www.safeworkaustralia.gov.au/sites/swa/model-whs-laws/pages/model-whs-laws

21 identification and management of fatigue risk factors; driver readiness, health and competence on the day; use of relief drivers and sub-contractors; and driver involvement and flexibility in the trip schedule.

An independent evaluation of the Fatigue Management Program (FMP) involved driver surveys in participating and non-participating (control) companies before and after the Program began (Burgess-Limerick and Bowen-Rotsaert, 2002). The evaluation found that drivers working under FMP conditions were less prone to fatigue risk than those not working under these conditions.

The National Road Transport Commission – now National Transport Commission (NTC) – also worked together with State road authorities to pilot an accreditation scheme in mass management in Victoria and another pilot in maintenance management in New South Wales.

The apparent successful implementation of the initial pilot alternative compliance projects led to the development of a national accreditation scheme (Baas and Taramoeroa, 2008). At their meeting in late 1997, Australian Transport Ministers approved the national framework for ‘Alternative Compliance’ schemes and agreed to offer industry a scheme that became known as the National Heavy Vehicle Accreditation Scheme (NHVAS). The NHVAS is now administered through the National Heavy Vehicle Regulator (NHVR).

1.6.1 Alternative compliance or ‘government concessional’ schemes The intention of the NHVAS was to provide incentives for heavy vehicle transport companies to become more proactive in safety management beyond complying with transport regulations. Concessions to enhance operational efficiencies were offered to NHVAS accredited companies.

In November 2007, the Australian Transport Council voted unanimously to approve amendments to the Model Heavy Vehicle Driver Fatigue law, the NHVAS Business Rules and a common implementation date for the reform of 29 September 2008.

Now the NHVAS has modules on mass management, vehicle maintenance and fatigue management.7 However, except for some transport operators based in Western Australia (who must be accredited under the Western Australia Heavy vehicle Accreditation Scheme (WAHVAS)), the choice of whether or not to become accredited under any or all of the modules of the NHVAS is at the discretion of the transport companies.

By accreditation through the NHVAS, the Government aims to allow transport operators to demonstrate compliance with aspects of road transport law through an auditable management system. In return for this, these operators are subject to lesser intensity of conventional

7 See https://www.nhvr.gov.au/safety-accreditation-compliance/national-heavy-vehicle-accreditation-scheme

enforcement. For example, companies that are accredited to the maintenance management component are exempt from annual inspections of their vehicles. Incentives for operators accredited to mass management have access to higher mass limits for vehicles with road-friendly suspensions. Compliance with NHVAS accreditation requirements is audited by external auditors, drawn from RABQSA8, however companies can choose among these auditors the ones they engage to audit their operations, raising questions of independence on the part of these auditors.

The Government also allows operators accredited to these and some industry schemes as a “reasonable steps” defence in workplace safety court cases brought against them, despite little peer-reviewed scientific evidence to show that accreditation makes them safer operators. The reason for doing this thesis study was to gather the evidence about what works and what does not.

1.6.2 Industry safety management accreditation schemes The two main auditable safety management programs developed and managed by the road transport and logistics industry are TruckSafe, managed by the Australian Trucking Association, and the National Logistics Safety Code, managed by the Australian Logistics Council.

1.6.2.1 TruckSafe TruckSafe (Australian Trucking Association, 2014) is a business and risk management system aimed at improving the safety and professionalism of trucking operators. It is an industry initiative, which aims to deliver business advantages to accredited operators. TruckSafe accreditation is based on four key standards:

Management - Aimed at ensuring that a trucking operator has a documented business system which covers each of the standards;

Maintenance - Aimed at ensuring vehicles and trailers are kept in a safe and roadworthy condition. This standard covers the requirements for daily checks, fault reporting and recording, fault repair, scheduled maintenance, maintenance records and documentation, maintenance responsibilities, internal review, and maintenance training and education. TruckSafe maintenance also complies with NHVAS maintenance standards;

Workplace and Driver Health - Aimed at ensuring that drivers are fit and healthy and OHS requirements are met. This standard covers requirements for workplace health and safety, driver health screening (including medicals), the role of the medical practitioner, rehabilitation and fatigue management; and

Training - Aimed at ensuring that drivers are licensed, authorized and trained for the tasks,

8 RABQSA is an Australian personnel and training certification body. RABQSA was created in 2004 from the acquisition of the personnel certification activities US-based Registrar Accreditation Board (RAB) by Australia-based Quality Society of Australasia (QSA).

23 which they are undertaking.

These are the minimum standards a trucking business are required meet in order to be accredited by TruckSafe. Compliance with TruckSafe standards is audited by external auditors, drawn from RABQSA. Unlike the case with NHVAS, auditors are selected by the TruckSafe Board, providing somewhat more independence of the auditors. However, there are no requirements regarding pay methods and other management practices that have evidence supporting their importance.

For operators, accreditation may show that they are meeting their due diligence and duty of care obligations, in regards to the items listed above, and accreditation has been acknowledged by the Victorian Government and the National Heavy Vehicle Regulator as an industry code of conduct that can be used as evidence of reasonable steps in court cases brought against them. However, the peer-reviewed scientific evidence to support this assumption is weak. No independent scientific evaluations of these schemes have been undertaken.

1.6.2.2 Retail Logistics Safety Code (RLSC) The RLSC began in 2006 with five members – retail majors Woolworths, Coles and Metcash, and transport giants Toll and Linfox. It has grown to include more than 60 businesses involved in the retail logistics supply chain. The Australian Food and Grocery Council, the National Transport Commission and the Australian Trucking association have also been centrally involved in the Code’s development. The Australian Logistics Council (ALC) is the custodian of the Code 9 (meaning that they directly administer the scheme and determine compliance with the Code).

The ten-point code aims to support a clear chain of responsibility in freight logistics, which involves all parties in the operation from the supplier and retailer to the carrier and logistics provider. In setting operational and administrative guidelines for compliance with the spirit and letter of law, the code recognises the importance of public safety and amenity in retail logistics operations.

While the Code is voluntary and applies only to signatories, it is a stated requirement of doing business with the major retailers. It aims to reinforce minimum levels of safe practices to help those in the retail logistics supply chain manage their obligations under relevant road transport and workplace health and safety laws.

The Code of Practice is made up of 3 parts:

a 10-point Code of Conduct document (which is signed by the company and ALC) which is the formal commitment (see Appendix A); a document which sets out operational and administrative guidelines; and a responsibility matrix which details the responsibilities for each specific role in the supply chain.

9 See http://austlogistics.com.au/

The RLSC Code of Practice requires all signatories to undertake annual compliance audits conducted by an RABQSA approved auditor. However, while the intention of the Code is to enable parties in the transport chain to come together and jointly review audit findings, termed Partner Audit Review (PAR), this is not occurring in the program as intended (Driver, 2015). A recent review of the PAR process found that this may be due to:

Prolonged time frame between audits and the actual Partner Audit Review (PAR), Review of each audit report with both parties in attendance takes too long, one day, If parties require their auditors to attend the PAR there is a significant additional cost, Requirement to conduct a PAR is not an NLSC question, Audits are closed out before the PAR is completed, Reluctance by some code signatories to participate in the process, and Perception that a master slave relationship exists and as such that some parties are unwilling to challenge the other for fear of losing a contract (Driver, 2015, p.11).

Despite these weaknesses, in September 2011 the RLSC was registered as a Code of Practice under Victoria’s Road Safety Act 1986. Registration of the Code means that, as for other industry schemes, if a company is charged with a breach under one of the provisions of the Act, it can use the Code to provide the basis for a ‘reasonable steps defence’ under this and the National Heavy Vehicle Law.

The ALC now has a generic National Logistics Safety Code and has been tailoring it to suit specific industry sectors, including steel, coal seam gas, electrical cable, and tanker industries. However, this Code has not been independently evaluated for its efficacy. So, there is no scientific evidence that it makes a contribution to improved safety outcomes.

1.6.3 Relationships between accreditation and safety outcomes There are mixed assessments of safety accreditation schemes. Despite the perceived benefits of these schemes there is no peer-reviewed scientific evidence that accreditation by either the Government concessional programs or the industry schemes achieve improved safety outcomes (Mooren and Grzebieta, 2011; NTC, 2009). Perhaps this is because there have been few attempts to evaluate them. One study of truck defects by NHVAS accredited operators and non-accredited operators, commissioned by the NSW Roads and Traffic Authority, found that “there was no significant difference in the rate of major defects in vehicles involved in alternative compliance schemes compared to those not in such schemes even though the proportion of vehicles involved in alternative compliance schemes increased from 9% to 16%” (Jansen, 2009, p.5).

Another study into the effectiveness of NHVAS and TruckSafe found that vehicles accredited under these schemes are, on average, significantly safer than vehicles that were not accredited (Baas and Taramoeroa, 2008; Rufford and Baas, 2006). The authors calculated that the difference in average

25 crash rates was substantial, with vehicles accredited to the schemes having between 1/2 and 3/4 fewer crashes per year on average than non-accredited vehicles. However, no direction of the association of accreditation and safety outcomes was established. The authors concluded that there is some evidence of improved safety performance but that the idea that improvements resulted from the process of accreditation must be treated with some caution. Moreover, the work by Baas and Taraoeroa was a consultancy project that was not peer reviewed. These authors admitted “very little research has been undertaken on how company management influences driver behaviour and road safety” (Baas and Taramoeroa, 2008, p.36). One major insurer 10 offers premium discounts to TruckSafe accredited companies, after finding that these companies have roughly half the claim rates of non-accredited companies. However, it has not been established that the accreditation made these companies safer or whether they would be operating as safely without being accredited. These causal links have not been established (Mooren and Grzebieta, 2011). It may be that companies that obtain accreditation would have had fewer crashes than those that do not, even if they had not become accredited.

A review of NHVAS in 2009 concluded that “it cannot unequivocally be said that the policy has improved road safety” and that rather than being an alternative compliance program it is more aptly described as a national permitting policy (NTC, 2009, p.12). Indeed, as the concessions offered include letting drivers work up to 14 hours per day, they may be putting these drivers at higher risk of becoming fatigued. Similarly, concessions including fewer roadside inspections could be resulting in more overloaded or poorly maintained trucks traveling on roads.

TruckSafe has a clearer focus on safety management (with compulsory safety management modules) than does the NHVAS (NTC, 2009) and from an insurer’s perspective 11 accredited companies are 50-75% safer (Baas and Taramoeroa, 2008). However, neither TruckSafe nor the National Logistics Safety Code have been shown to increase the level of safety through participation in these programs. Also, as there are few Government incentives for these schemes, the industry uptake is low (~10% for TruckSafe and less for NLSC12).

In addition, TruckSafe and NHVAS accredited companies have been prosecuted for serious breaches of Australian regulations, including the tampering of their trucks’ mandatory speed limiter devices (ATN, 2012)13.

Hence, this thesis aimed to identify safety management practices that show evidence of distinguishing between heavy vehicle operators with better and poorer safety outcomes.

10 See https://www.nti.com.au/files/case-studies/NTI-case-study-3.pdf 11 To an insurer, it does not matter whether the companies are safer as a result of accreditation, or they are accredited because they are safer. 12 According to verbal advice from the Australian Trucking Association and the Australian Logistics Council 13 NSW Police advised that Lennon’s Transport, charged for tampering with speed limiters was accredited to TruckSafe and NHVAS.

1.7 Constraints on regulatory effectiveness

From the perspective of public safety and workplace safety, injuries and fatalities, as well as economic costs due to incidents that occur in the road freight transport industry, present a very substantial problem and warrant attention. Trucks are overrepresented in serious crash statistics relative to the general road crash rates and truck drivers are overrepresented in work related fatality rates.

Despite government efforts in heavy vehicle safety regulation and enforcement efforts in Australia, there are deficiencies and limitations to their effectiveness. This is evidenced by the continuing high incidence of regulatory breaches being detected (Gardner, 2013) as well as high crash and injury rates that do not seem to be abating.

Until Chain of Responsibility (COR) principles were introduced in Australia, the heavy vehicle regulatory enforcement targeted only drivers for driving breaches and vehicle defects. Since CoR principles were brought into transport regulations, police are now able to issue notices of breaches to transport companies and others in the logistics chain that have influenced an unsafe heavy vehicle action. Police forces across Australia have been making concerted efforts to enforce COR provisions (McKay, 2011). However, it is still much easier for police to issue a breach notice on only the driver and enforcement only focuses on drivers (O'Neill, 2014).

Work safety regulations can be used to prosecute trucking companies as employers, but the perception of many managers of these companies is that they cannot be held accountable for driver errors and violations on the road. Indeed comments by some of the managers in Study 3, reported in Chapter 5, indicate that they believe that safety on the road is the driver’s responsibility. This weakness, together with the ongoing police tendency of issuing citations only to drivers (O'Neill, 2014), may have resulted in employers not accepting responsibility for safety in the public road environment.

Moreover, the Australian Government’s concessionary schemes (NHVAS) offer trade-offs on efficiency and safety with no clear demonstrated net safety benefit. The industry accreditation schemes like TruckSafe and the National Logistics Safety Code have been credited with association with lower crash rates, but these programs have not been evaluated in a scientific manner that would support these claims with sound evidence.

1.8 Conclusions

Despite the efforts of Australian Government agencies and industry organisations, serious safety problems in the trucking industry persist. There appear to be some substantial, challenging, industry characteristics that perpetuate high levels of safety risk associated with the industry. The bulk of road freight transport buying is concentrated into a small number of large companies, and yet the vast

27 majority of road transport providers are small operators. The intensely competitive environment, low profit margins and tendencies to contract out driving tasks to those with little bargaining power result in safety vulnerabilities like self-imposed time pressures on drivers, speeding, driving whilst fatigued and using stimulant drugs. Transport legislation, particularly the Chain of Responsibility provisions, contain weaknesses that result in difficulties in enforcement and prosecution, making safety regulation of the industry and transport operations less than effective.

Accreditation to industry auditable safety management schemes, such as TruckSafe and the National Logistics Safety Code, may be helpful, but scientific evidence that has undergone appropriate peer review of their effectiveness is non-existent. They are yet to be evaluated for their effectiveness through a scientifically rigorous evidence base that demonstrates ability to reduce crashes and other safety incidents. Moreover, the uptake of these programs is quite low.

This Chapter described the road freight transport industry as one of the highest risk industries and how heavy truck driving represents markedly higher safety risk compared to other road users. The analysis of inherent risks in this industry suggests that this higher safety risk is not due simply to high levels of exposure to the road environment but also to characteristics of the task of heavy vehicle drivers and the structure of the industry. In addition, perhaps a lack of effective of transport and workplace safety regulatory enforcement has failed to keep safety of heavy vehicle freight operations in check. The review of approaches put in place in this industry in Australia to attempt to improve safety for heavy truck driving, such as regulation, accreditation and compliance, shows little evidence that they have been successful.

To some extent this is because no formal scientific evaluation has been attempted, but where the approaches have been tested, they have been shown to be wanting. As the road freight industry is predicted to increase markedly over coming decades, the need to obtain better control of safety in the industry is even more important. It is clear that new approaches are needed to address the safety risk in heavy trucking.

The aim of this Chapter was to explore the current approaches to public safety on the road, and safety in the workplace, as these areas are most relevant to road freight transport safety. The review critically examined the evidence in order to identify approaches that are likely to be effective for heavy trucking. The conclusion is that stronger scientific evidence-based safety management strategies are needed for the heavy vehicle transport industry. Currently there is no such evidence-based safety management system suitable for improving safety in the trucking industry.

The proposition of this thesis is that an efficacious safety management intervention can be devised by identifying the characteristics that distinguish companies with good safety outcomes from companies with poorer safety outcomes. A proxy measure of safety outcomes was used. Insurance claim rates per truck was used for recruiting better and poorer performing companies for investigation of safety management characteristics that distinguish between them. Insurance claims mainly include

the range of safety-related outcomes resulting from incidents and crashes as well as other unexpected incidents, many of which may impinge on safety. The characteristics that distinguish higher and lower claiming companies can then be used to build a safety management system that would be expected to result in improved safety outcomes in terms of incidents and injuries. That is, it was assumed that companies with higher truck insurance claim rates operate with a higher risk of incidents and injuries.

Thus, the overall aim of the research was to identify a system of safety management that would be likely to produce good safety outcomes. Two specific objectives of this research were to:

1. identify the distinguishing characteristics and practices of heavy vehicle transport operating companies with good safety records and those with poorer safety records; and 2. develop a evidence-based safety management system suitable for companies that operate heavy vehicles for transport of goods that will achieve good safety outcomes.

1.9 Thesis structure and content

This Chapter has presented the case for developing a safety management system aimed at improving safety in the heavy vehicle transport industry. Chapter 2 reports findings in theoretical research literature that provide a scientific framework, based on peer-reviewed research for the construction of a safety management system (SMS). Then Chapter 3 presents findings of a strategic literature review (Study 1) that forms a base of what is already known about safety management, again based on peer-reviewed evidence. Chapter 4 presents the findings of Study 2, a cross sectional survey of managers of groups of trucking companies distilling differences in management characteristics between those with better safety performance and those with poorer safety performances. After this, the third study, an in-depth investigation of a sample of Study 2 candidates aiming to validate Study 2 findings will be presented, offering additional insights into important safety management characteristics in the trucking industry. These three studies were triangulated and their findings synthesised. This is presented in Chapter 5. The analysis is then used as the basis for development of a safety management system for the heavy vehicle transport industry and is described in Chapter 6 together with conclusions of this thesis.

29 Chapter 2: Current Approaches to Safety Management

In order to determine what a safety management system is for companies that operate heavy vehicles, it is necessary first to establish what current approaches are to safety management, and what is the current scientific evidence that underpins them.

This chapter first reviews the definitions and models of safety management. The traditional and existing approaches to road safety are then described and compared with other models, particularly workplace safety models. The aim is to assess the strengths of comparative theories and approaches, and more importantly the evidence that underlies the ways that safety is managed.

2.1 What is safety management?

As the aim of this research is to develop a safety management system that reduces the risk of crashes, it is important to define what is meant by safety management system. A simple definition for safety is the absence of unacceptable harm (Hollnagel et al., 2007). Svanström (2000) defined safety as “a state characterised by adequate control of physical, material or moral threats which contributes to a perception of being sheltered from danger.”

In the context of safety management the focus is on non-intentional or accidental harm. Accidents can be defined as hazards “materialising in a sudden, probabilistic event (or chains of events) with adverse consequences (injuries)” (Hovden et al., 2010). Prevention of harmful events is the goal of safety management. The way safety is measured tends to be a quantification of the number of harmful events. The categories of things to quantify include outcomes like damage or loss, incidents, hazardous conditions and unsafe acts (Kjellén, 2000). These are then translated as the level of failure of systems to manage safety, or failure of safety management.

In road safety, the term “accident” has largely been replaced by “crash” or “incident”. This is because the word accident carries connotations of a random, unpredictable and therefore an unpreventable event. The term “accident” is all but absent from the World Report on Road Traffic Injury Prevention for this reason (Peden et al., 2004, p.7). In this thesis the terms, accident, incident and crash, are all used because the workplace safety and road safety literature use the terms, all in the context of prevention of harmful or injurious events.

2.2 Road safety history and the development of scientific approaches

Nearly a century after the first motor vehicle road fatality occurred, Dean (1947, reprinted 2007) wrote a book entitled, “Murder Most Foul: a study of the road deaths problem.” The author concluded that “the ‘reconstruction of Britain’ will indeed be a dismal failure if it includes as a permanent feature of the national life the killing and maiming of a quarter of a million, or more, of persons every year on the roads…there is no reason for failure…all that is needed is the will to act” (Dean, 1947, reprinted 2007, p.114).

30 Writing during the post World War II era, Dean believed that increases in road deaths were directly related to the rise of fascism, pointing to the fact that Nazi Germany and Mussolini's Italy had the highest per vehicle rate of road fatalities in 1934. He explained that, in these countries, the motor interests were the biggest supporters of Hitler and Mussolini. Dean illustrated how motor interests were protected in all road safety efforts, by targeting the behaviour of vulnerable road users through education and punitive actions alone. Typical Nazi Government responses were to introduce fines, collectable on the spot for "careless walking" and for "endangering traffic" [while walking in the road], and for riding a bicycle two abreast. Dean lamented that Britain was also influenced strongly by motor interests, citing many examples of media comments in the mid-1930s about the imposition of a speed limit - that restrictions on speed would "fatally damage the motor industry."

Initially the response to road deaths was to try to educate the community about how to use the road properly on the assumption that the problem was a knowledge and skill deficit amongst drivers - to drive safely - and to pedestrians and pedal cyclists - to keep out of the way of motor vehicles. Gradually training, testing and licensing processes emerged in countries where motor vehicle use was growing. Later some focus was directed to the motor vehicles themselves, but this was not very effective until American legal specialists, led by Ralph Nader, took on the car industry with law suits focusing on the intrinsic unacceptable unsafe features of cars. In 1965 and 1966, public pressure grew in the US to increase the safety of cars, culminating in Ralph Nader's book "Unsafe at Any Speed: the designed-in dangers of the American automobile" (1 972).

In 1966, the US Congress held a series of highly publicized hearings about highway safety, passed legislation to make installation of seat belts mandatory, and created several agencies that would eventually become the National Highway Traffic Safety Administration (NHTSA)(Ward et al., 2006)

At the same time, the 'scientific method' of analysing road injury causation was embraced in the 1960s following the work of Dr William Haddon, an injury epidemiologist. Haddon's injury analysis method called for identification of contributing factors prior to, during and after the crash event, grouping them into three categories, vehicle, road environment and human factors (Haddon, 1968). That is, in a manner similar to traditional medical epidemiological examination of how infectious diseases occur, injury epidemiology conceptualised the components of a host (human), a vector or agent (vehicle) and an environment (road) to find how the transfer (in this case of harmfu l energy forces) would result in injury. Figure 2.1 presents the 2-dimensional risk factor assessment matrix developed by Haddon.

Road User Vehicle Road environment Pre-crash Crash Post-crash Figure 2.1 Haddon Matrix for identifying injury factors

31 This injury analysis model is fairly simple to use in road safety. It is a matter of looking at, for example, road fatality data and counting the number of times a behavioural, vehicle and/or a road environment factor was present, before during and after a crash, and putting these numbers in the appropriate cells of the matrix. Then, by adding all of the cases where road user factors were present and dividing by the total number of crashes in the data set, a percentage of crashes involving behavioural factors can be calculated, and so on for vehicle and road environment factors. This analysis provides information about the relative contributions of road injury factors, thus enabling a focus on investments in prevention programs to address those risk factors that occur most often in fatal crashes.

This was the springboard for a more systematic analysis of road injury. Injury specialists, particularly in the Western world, began to adopt this method with bio-mechanical engineers looking at the most prevalent vehicle features that contributed to more and more severe injuries, civil engineers examining the most prevalent road environment features, and behavioural scientists investigating the most prevalent unsafe road behaviours contributing to crashes and injuries. The Haddon approach thus enabled road safety practitioners to identify clusters of road injury factors within the matrix that could then be investigated further with social, behavioural and engineering research methods. Such analysis of data on the contributing factors to, and circumstances of, crashes formed the basis of the then prevailing scientific approach to road safety management.

However, observing the data clusters within the cells of the Haddon Matrix can be misleading if the interrelationships between and among the various factors are not understood. Moreover, the Matrix does not serve as a tool for explaining what caused the injury factors to manifest, nor does it enable the injury analyst to know what factors are most important to address. In other words, it provides only snapshots of the components that contribute to injury. While Haddon speaks of three phases of injury causation and a multiplicity of contributing factors, his model does not recognise the dynamic and compounding nature of injury causation processes; nor does it convey the interrelationships among the cells of the matrix. Instead, injury factors appear in the Matrix as discrete occurrences. Individual road crashes, particularly fatalities, may be reported with detailed chronologies of events but the data are then dissected and filed in categories where groupings of contributing factors are aggregated in human, road and vehicle sections. Moreover, the chronology of events is usually limited to what the road crash investigator can find at the scene of the crash. It is rare, for example, that police collect data regarding purpose of journey.

Studies using the Haddon Matrix (Morgan et al., 1999; Sabey and Taylor, 1980) began to show that human factors were more prevalent than other types of factors. These studies showed that over 90% of fatal crashes involved human behavioural factors (mistakes or violations), 28% find that road environment factors are involved, and 8% involve vehicle factors. In practice there are usually a number of factors that combine to result in road crash outcomes, which is why the sum of the percentages exceeds 100%.

The prominence of human behavioural factors found by this method of analysis suggests that the largest share of the road safety intervention is best devoted to addressing road injury risks associated with human

behaviour. Some effective methods of deterring unsafe road behaviour through law enforcement, combined with public education, have been carried out with good effect (Pilkington and Kinra, 2005), and extended training and licensing of young drivers have helped to reduce their involvement in injury crashes (Senserrick and Whelan, 2003). However, it is arguable that fixing vehicle and road environment risk factors has a more lasting effect in reducing these crashes than education and enforcement, or at least making crashes less harmful.

Along with the two-dimensional epidemiological model of injury analysis, Haddon developed ten basic strategies for injury prevention that can apply across all injury categories. The concepts encompass primary, secondary and tertiary actions that can be taken to prevent harm by injury-causing events. Another important concept in injury prevention highlighted by Haddon’s work is passive versus active safety (Haddon, 1980). Active safety occurs when the human host does something to prevent or ameliorate harm. Passive safety occurs when conditions are designed to prevent human errors becoming harmful or injurious. In other words, passive safety does not rely on humans to make decisions in favour of safety and compensates or corrects for their mistakes. The use of the Haddon injury analysis tool tends to treat all contributing factors equally and typically finds that in over 90% of crashes, human factors are involved; but these factors can be addressed through passive system measures as well as with active behavioural management measures.

In this and other dimensions, the Haddon model does not enable ready analysis of the combined effects of road safety countermeasures being implemented simultaneously (which is often the case in jurisdictions with strategic road safety programs). Elvik (2009) concludes that “there is a need for more research in this area in order to develop more sophisticated models for estimating the combined effects of road safety measures (p.880).”

Moreover this model is premised on the expectation that there will always be road deaths, and that fatality risk can only be reduced, not eliminated. Under the Haddon approach, risk factors and countermeasures are systematically identified, but often a cost-benefit analysis is used to guide decisions about which countermeasures would be implemented and how far they would go to restrict individual mobility or transport efficiencies (Grzebieta, 2013). For example, road design engineers still accept that 15-20% of people will not recover from an injury crash that results from insufficient clear zones on roadsides (AustRoads, 2010).

By the mid-1990s, Sweden (with “Vision Zero” (Swedish Ministry of Transport and Communications, 1997; Tingvall, 1998)) and the Netherlands (with “Sustainable Safety” (van Schagen and Janssen, 2000; Wegman, 2005)) both introduced a new dimension to road safety – the ethical dimension. Both reflected the principle that trade-offs between safety and mobility are not normally acceptable. As modern society generally expects workplaces to be safeguarded from injury risks, these policies set an expectation that road users will not be harmed in the public road and traffic environment. The policies shift the primary responsibility for safety from the road user to the road designer. They are formally based on the premise

33 that people are prone to making mistakes. Both approaches are based on the view that the road and traffic system needs to be fortified to stop road user mistakes becoming fatal. When introducing the Vision Zero bill into the Swedish Parliament, the proponent advised “It is true that 95% of all crashes or collisions depend on human error, but according to Vision Zero philosophy, 95% of the solutions are in changing roads, streets or vehicles” (Johansson, 2009, p. 827).

The Australian “safe system” approach grew out of this principle (Tingvall and Haworth, 1999). It has been claimed that this approach represents a “paradigm shift” in road safety approaches (Grzebieta, 2001; OECD, 2008; Rechnitzer and Grzebieta, 1999). Traditionally, the method of identifying road injury contributing factors has been to identify all human, road and vehicle factors that are present before, during and after the road injury event (crash). Then counting the number of times each individual factor is found to be present in a number of injury crashes, say all injury crashes in a 12-month period, determines the prevalence of each factor. Using this approach, all road injury factors are given equal weighting. Defining prevalent factors as human, environmental or vehicle tends to mask the systemic underlying influences and interactions between the factors. So, for example, speeding is found to be a human behaviour quite often contributing to injury crashes. Not taking into account the systemic influences on speeding behaviour can lead to interventions to directly change this behaviour of road users such as education and enforcement. In contrast, systems thinking would lead to investigating system influences on the behaviour. These could include the design of the road, the characteristics of the vehicle, and socio-economic factors such as time pressure. The paradigm shift from the traditional Haddon approach to safe system road safety is from treating factors that contribute to road injury as notionally equal to all other injury factors, to a greater priority given to systemic risk factors. Rather than assuming that there will always be injury risks inherent in road travel, the safe system thinking conceptualises and pursues the development and management of a road traffic transport system that is inherently safe for human users. The safe system approach calls for road, traffic and vehicle design parameters that acknowledge human fallibility and vulnerability. It places a biomechanical injury tolerance criterion and consideration of human fallibility as the central governing principles underpinning any road safety policy decisions and system development. In other words, if a crash occurs its severity should be at a level that the road user is not injured or at least is likely to recover from their injuries.

System thinking is not new in the safety field. It just has not been fully applied in road safety. Indeed, nearly 20 years ago Rasmussen (1997) reviewed a range of models for understanding accident causation and emphasised the need for a multi-disciplinary understanding that human activity occur within a dynamic socio-technical system of interactions. Risks and risk management occur at a number of levels within this system, and various research disciplines can contribute to a richer understanding of each level of decisions and actions. The judgements and actions of people at every level of the system can either trigger an accident process or divert the accident process. These judgements are influenced in a dynamic process of socio-technical evolution. For example, at the political government level changes in public

attitudes can shift government commitment to safety measures. Economic recessions can increase and decrease risk exposure. Competencies of managers and workers as well as changes in technologies can influence safety decisions. Figure 2.2 depicts these levels and influences.

Figure 2.2 Socio-technical system involved in risk management (Rasmussen, 1997, p.185)

Dekker (2011) is sceptical about the likelihood that safety controls can be fully effective given the level of complexity of the modern world. He identified characteristics of complex socio-technical systems, such as technologies that cannot be fully controlled, social processes and environmental pressures, that lead gradually, but inevitably, to a “drift into failure” of safety management. Salmon et al (2012) argued that the road transport system is a complex system with key “failure” characteristics. For example, road transport systems are open to unpredictable elements (such as fuel prices, road user behaviour, etc.), road designers and road users not understanding all components of the system, nor interactions within the system, and government authorities not keeping pace with regulatory needs associated with rapid advances in vehicle technologies. Salmon et al say that an increasing understanding of ideas such as drift

35 into failure, combined with a shift in thinking towards the road transport system in terms of its complexity and socio-technological influences are needed to manage road safety in a more effective way.

Recently, the search for new, better approaches to road safety led to a review by Hughes et al (2015) of safety models that could be used in road safety. The authors concluded that a systems approaches have the potential to be effectively used in road safety on the basis that they have been applied successfully in other fields. However, there is little evidence that a systems approach has been applied in practice to road safety relative to how it is applied in other safety domains. Moreover, it is argued that the current Australian road safety strategies (ATC, 2004) and the Global Plan for the Decade of Action (WHO, 2010) are not consistent with system thinking while claiming that they are underpinned by safe system principles (Salmon and Lenné, 2015). The strategies still contain countermeasures mainly within the components of the system (road, vehicle and human user), rather than on underlying systemic factors, such as economic pressures on the road freight industry that provide incentives for truck drivers to break road safety laws, poor land use planning resulting in speed zoning that increases pedestrian safety risks, and energy providers erecting dangerous power poles making the road environment more lethal in the event of a crash.

Finally, regardless of the model or approach taken, important aspects of effectively managing road safety require evidence-based interventions, political will to implement them, collaborative institutional arrangements, and community demand for safety (OECD, 2008). It could be argued that Australia has been successful in influencing driving its road safety strategies with a strong evidence-base, reasonable political will and collaborative institutional arrangements providing good synergies for effective implementation of road safety strategies. However, Johnston (2010) argues that creating a community demand for a safe road and traffic system in Australia, where the prevailing car use culture does not align with the primacy of safety, is the dimension that is proving a challenge.

2.3 Applying road safety approaches to work related driving

Currently, road safety and fleet safety research still largely concentrates on the Haddon type of analysis of road injury causation factors (Mooren et al., 2009a). Some models have been developed around work- related driving to address some of the issues specific to driving in this context.

The first international conference on “Road Safety at Work” hosted by the US National Institute of Occupational Safety and Health, in February, 2009, brought together road safety and fleet safety practitioners. The conference “white paper” concludes that:

Overall, the Haddon Matrix ... provides an all-encompassing systems-based framework that incorporates all the good practice identified in the report, and is supported by a small, but increasing body of published evaluation data, and many as yet unpublished outcomes. (Murray et al., 2009b, p. 60).

Dubens and Murray proposed a modified Haddon Matrix. Figure 2.3 depicts sample corporate interventions that are fitted with the usual cells used in road safety, addressing people, road and vehicle before, during and after a crash, plus some additional columns for management culture, journey and external conditions. They advised that “the percentage figures in the headings are shown for discussion purposes, and represent a best estimate of the relative importance of each factor from a good practice perspective (Murray et al., 2009b, p. 54).” Thus, the example chart indicates where an organisation should devote resources, programs or actions.

Figure 2.3 Haddon Matrix for Fleet Safety Management (Dubens and Murray, 2009, p. 16)

According to Dubens and Murray, companies have used this framework to guide them in determining what specific aspects of safety management would need to be addressed to assure good control and mitigation of fleet related workplace risk (Murray et al., 2009a; Murray et al., 2012). Case studies (Bradley, 2014; British Telecom, 2010; Murray et al., 2009a) demonstrate how the Haddon Matrix has been applied to help audit safety performance, identify gaps, develop business cases and programs, evaluate performance and structure investigations after major collisions. Murray’s use of the Haddon Matrix for indicating management actions pre-crash, during crash and post-crash, is a way of conveying things that could be done to better prepare organisations to manage work related driving risk. However, in examining the case studies it is unclear how the Haddon based tool has played a part in safety improvement achievements. Moreover, while the Matrix may provide a systematic method of examining and addressing risk factors, this framework does not reflect a dynamic systemic approach to understanding the work related driving risk environment. That is, it lists groups of countermeasures that can be chosen for implementation as discrete safety management actions, rather than treating the work related driving process as an activity that occurs in a complex ever-changing environment. The problem with this approach is that, because it

37 does not recognise the mediating influences of some countermeasures on others it can miss some important opportunities. For example, perceived management commitment to safety moderates effectiveness of incentives, safety reporting, empowerment of employees (Al-Refaie, 2013). So, if an employer chooses to implement a safety reporting system without having a good safety culture in the company, the reporting system may not be as effective as expected. Finally, the model has not been evaluated and hence there is no evidence that it does result in reduced crash outcomes.

Runyan (1998) supported the use of the Haddon Matrix, saying that it is a useful tool in determining actions to guide decisions about interventions for improving work-related driving safety. However, she thought that the model could be enhanced with a third dimension to include additional decision-making criteria. This third dimension proscribes a range of resource management decision-making criteria. These criteria include effectiveness, cost, amount of freedom (of decision) acceptable to motor vehicle users, equity of interventions on users, whether interventions will stigmatise some groups unfairly, preferences of the group affected, feasibility of implementation, and other relevant criteria for decision-making. Figure 2.4 provides a description of Runyan’s 3-dimensional Haddon Matrix.

Figure 2.4 Runyan’s 3-dimensional Haddon Matrix (Runyan, 1998, p. 304)

Adding this dimension to a static causal model of injury factor categorisation adds more boxes to work within, giving suggestions of the intervention decision-making criteria to apply. But it does not fix the inherent limitations of the model itself as an injury intervention tool. In addition, it seems to add a complexity that may be too cumbersome to use as a tool for planning and programming safety

interventions (Mooren et al., 2009a). Again, there is no evidence that it is effective in reducing incidents or injury severity.

Also, taking a snapshot of the contributing human, vehicle, road and social factors, then implementing actions, based on subjective judgment about the things that should be logically done through an organisational decision-making process, only takes a piecemeal approach to interventions. The Haddon Matrix is limited to placing identified injury factors into categories to determine the most prevalent factors. But adding more boxes for decision-making factors contributes little to sound decision-making about ways to intervene in the injury causation process. The 3-dimensional model is still a static piecemeal approach to injury prevention that does not convey the interactions between the cells of the Matrix. Moreover, a search for case studies using the Runyan 3-dimensional Haddon model did not find any evidence that it has been used in the work-related driving context.

Increasingly researchers are identifying limitations to the utility of the Haddon approach in improving work- related driving safety. Stuckey et al (2007) determined that the traditional models used to date in road safety were not sufficient for the analysis of risks associated with occupational light vehicles (OLV) and developed what they called an “integrated systems model” that recognises that road risks emerge within a complex system of interactions among “multiple levels of influence on outcomes” (crashes and injuries). As shown in Figure 2.5, the individual road user (the “locus of injury”) is placed in the contexts of immediate and external environments (vehicles and roads), organisational environments (prescribing travel tasks), and broader social, legal and economic environments (ethical and regulatory). Recognising these environments acknowledges more aspects of the system in which driving is done and so provides for a systematic analysis of crash and injury determinants of risk exposure for occupational light vehicle risk.

39 Policy environment External influences: Relevant local, national and international public policy

Locus of injuries/fatalities: Drivers/passengers

Figure 2.5 Occupational Light Vehicle (OLV) Systems Model (Stuckey et al., 2007, p. 1008)

In this model, the elements of road injury risk are understood within a framework of immediate and increasingly proximal conditions and environments. The human, vehicle and environment factors from the Haddon matrix are replaced in Stuckey et al by a systems model that shows spheres of risk. The authors suggest that more can be learned about systemic factors that contribute to providing hazardous conditions that influence the final triggers to the injury event, if the examination looks further than the proximal locus of injury events.

This concept may be a useful one to consider in that it places the harm of occupational light vehicle drivers at the centre of the model and examines the layers of risk environments to manage around this problem. While other studies have examined the influence of the external environmental (Walters and James, 2011), a search of literature (scientific and grey) did not locate evidence of any attempt to use the Stuckey systems model for a holistic process of determining injury causation factors in all of the spheres described. The use of the model has not been evaluated for its efficacy. Moreover, while using this model, potentially more information about the context in which driving takes place can be gained, the approach is still static in that it calls for a descriptive snapshot of factors that influence safety.

2.4 Safety management approaches in other sectors

As truck driving is an occupation, and crashes cause work related injuries, it is appropriate to examine theoretical developments in the workplace health and safety (WHS) literature to attempt to identify approaches to safety management that could be used for the heavy trucking industry that actually do red uce crashes and/or crash severity.

2.4.1 Models of WHS accident factor analysis Many models have been developed to describe the causes and circumstances of workplace accidents. Many of these have been developed to enable more systematic assessments of risk and safety management responses. One of the earliest developers was Heinrich who developed a safety risk assessment model, known as the Domino Theory (Heinrich, 1931). In this model accidents are explained as a chain of causes and effects, suggesting that a weakness in safety defences in the system of work can undermine the more proximal defences. Kjellen and Larsson (1981) developed a two-level model based on an assumption of a normal (accident-free) work process. For accident investigations they suggest looking chronologically for weaknesses in safety defences, analysed as deviations from the process, and then searching for determining factors that contributed to the deviations in the process.

One of the more influential theorists in WHS is James Reason. Starting from his early work on understanding human error (Reason, 1990), he later argued and provided evidence that human error is not a cause, but rather a consequence of organisational shortcomings (Reason, 1997a). This is not to say that human errors did not occur; rather it was a recognition that human errors are shaped by upstream workplace and organisational factors.

By looking at injury causation as management system failures, it is possible to devise organisational controls and defences against active failures becoming injury events. As Reason (2000) states, “We cannot change the human condition, but we can change the conditions under which humans work (p.786).” Work injury events are preceded by latent conditions, built into the management system representing holes in defences in accident trajectories. The active failure that triggers an accident is at the end of a set of these holes in the system defences which line up to allow a hazard to become a loss event. This is depicted in Figure 2.6 as chunks of Swiss cheese lined up in a row showing that a clear path through the holes in an organisation’s safety defence systems can result in losses.

Figure 2.6 Swiss Cheese model of accident trajectories (Reason, 1997b)

Some weaknesses or holes in the safety defence system may have existed for a long time and are more difficult to detect in day-to-day work processes. Others are created by active failures, for example, a guard

41 on a machine that is not secured into place. Some defences lose effectiveness as a result of changed work processes or conditions. For example, automation of production processes may reduce risks associated with heavy lifting but introduce new risks of limb injuries (where limbs can get injured by machines). Or, in an effort to tighten defence about one hazard, another hole in the defence system is created. For example, Australian rules for working at heights require using a 3-point safety harness, but if a petroleum tanker driver does not have access to a ladder nor 3-point harness s/he is still required to check that the top hatch on the tank is secured before starting their journey to prevent safety hazards created by leakage of petroleum while driving.

Reason’s contribution to thinking about how accidents occur was to revise thinking about causation from only examining immediate events to include examination of things happening much earlier in time. He also pointed out that these pre-existing factors may lay dormant or latent for a long time - even years. His important contribution was to expand thinking about the role of error in accidents to encompass the possibility that errors can occur due to the system itself.

The recognition of system complexity and the dynamic nature of work processes in many industries has led to development of models that do not assume linear or sequential chains of risk factors. Rasmussen (1997) argued that a system-oriented approach based on functional abstraction was needed rather than decomposition of risk factors. He said that in safety management, action sequences and occasional deviation in the form of human errors should be replaced by a model of behaviour-shaping mechanisms, including work system constraints, boundaries of acceptable performance, and subjective criteria guiding adaptation to change. Moreover, Rasmussen points out that the concept of “event” is elusive in the sense that the more specific its definition, the less likely it is to recur. Hence, this form of causal analysis is less valuable as a tool for prevention. He says that it is best to define classes of event that can then be moderated through other conditions being present. The analysis then relies on the ability to decompose processes and factors within a dynamic flow of activity.

Others have recognised further difficulties with complex systems. Perrow (1984) argued that these problems are virtually insurmountable. Following incidents such as the Three Mile Island nuclear meltdown accident in 1979 14, Perrow concluded that such accidents are destined to happen when the system is so highly complex. He said that these “normal accidents” or “system accidents” are inevitable in extremely complex systems such as those that characterise nuclear power plants. Consistent with his Natural Accidents Theory, he argued that as organisations grow in complexity, the prediction and prevention of accidents cannot be possible (Perrow, 1984). He said that, in some circumstances, system failures cannot all be anticipated and can result in a “normal accident”. Further support for the Natural Accidents Theory was provided in a study of petrochemical plants citing that despite intense efforts to prevent accidents, these accidents persisted at the same rates throughout the 1990s (Wolf, 2001).

14 More information about this accident can be found at https://en.wikipedia.org/wiki/Three_Mile_Island_accident#Lessons_learned

This notion was challenged by a theory that suggested redundancies can be built into complex systems and a ‘culture of reliability’ can achieve a “High Reliability Organisation” to mitigate the risks inherent in complex operational systems (Roberts, 1990). This means that it is possible to put in place safeguards that would back up any failures in other safeguards so that no accident trajectory will succeed. In this way the safety defence system can match the complexity of the work process.

Wieck et al (2008), however, highlighted the difficulty of implementing such a system. They argued that this would require an organisation to be preoccupied with the possibility of failure, reluctant to simplify interpretations of risk factors, sensitive to operational processes, committed to resilience, and able to underspecify structures or make them flexible enough to enable fluid adaptation. They say that it is important to be mindful of the ever-present possibility of failure in order to succeed reliably in safety.

The ideas have been taken up in some more recent models. For example, a ‘high reliability’ model was developed by the Aeronautics and Astronautics Department of the Massachusetts Institute of Technology (Leveson, 2004), taking into account the need for a more dynamic approach to accident analysis and control. Using the space shuttle programs as examples, Leveson et al. (2007) illustrated how building safety controls for making a complex and dynamic process resilient to accident risks requires an understanding that the systems involved in carrying out the work are fluid and feature continual interactions among people, social and organisational structures, engineering activities and physical components. Therefore they proposed a new accident model that calls for continuous controls of processes, instead of controls on individual system component failures. The Systems-Theoretic Accident Model and Process (STAMP) 15 recognises that “accidents occur when external disturbances, component failures, or dysfunctional interactions among system components are not adequately handled by the control system (Leveson, 2004, p. 250).” Unlike the Haddon model predominantly used in road safety, this type of systems based accident model has a stronger focus on understanding why the control system failed and why they made components unstable, than on immediate error factors. In other words, it asks how the conditions were developed such that an operator error could result in harm. In the STAMP model, “systems are viewed as interrelated components kept in a state of equilibrium by feedback loops of information and control (Leveson, 2004, p. 250).” As such it is a dynamic continual process of examination of the operations of safety control systems and changing internal and external environments. Safety management is defined as a continuous adaptive control task. However, the STAMP model has not been evaluated, hence lacks evidence of its effectiveness.

The idea that safety systems could become more resilient also emerged from the work of Hollnagel et al (Hollnagel et al., 2007). This was based on the recognition that work processes are often not stable. Hollnagel explains,

15 The systems-theoretic accident model and processes (STAMP) was developed in recognition of the inadequacy of understanding accident causation as a linear chain of events.

43 Resilience engineering is a paradigm for safety management that focuses on how to help people cope with complexity under pressure to achieve success. It strongly contrasts with what is typical today – a paradigm of tabulating error as if it were a thing, followed by interventions to reduce this count. A resilient organisation treats safety as a core value, not a commodity that can be counted (Hollnagel et al., 2007, p.6).

Currently the debate about this is unresolved, with neither the Natural Accidents Theorists nor the High Reliability Organisation Theorists providing clear evidence to support either theory. Safety researchers (Hollnagel, 2008; Hopkins, 2000) have questioned a number of the common concepts that underpin safety management such as the idea that hazard barriers can be erected and embedded in a work system to block accidents from happening. In fact, Reason (2000) cautions that a reliance on these barriers can lead to complacency that can in turn create hazards (e.g. “forgetting to be afraid”). Rather than treating workplace safety management as a set of components that can block injury events, a more dynamic understanding of work processes and work environments is needed. This would enable us to understand safety management as actions and continual adjustments instead of setting up a series of barriers to block processes leading to adverse events (Hollnagel, 2014).

Certainly, any analysis of risk factors in a complex organisation brings challenges. Standard methods of scientific analysis tend to break down systems and processes into components in order to understand linear processes that result in an outcome. However, decomposing a complex system would mislead the analyst as the whole is greater than the sum of its parts (Le Coze, 2008).

Moreover, linear cause-effect analyses are confounded in complex organisations because there are feedback loops that influence parts of the system in less predictable ways. Organisations with complex systems are dynamic, adaptive and transformative as internal processes emerge and/or external influences affect the way that decisions are made and actions are taken (Le Coze, 2008).

By focusing strongly on discrete risk factors using a linear sequential model, engineering, education, and enforcement solutions may miss important opportunities to address risk at a systemic level. Moreover, the linear approach does not offer a way to analyse the interaction between risk factors or safety solutions.

Important changes in the way work is processed over recent decades signal the need to re-examine the way safety is understood. New technologies and work methods have changed to the point where viewing safety outcomes as cause-effect relationships is no longer effective for the purpose of understanding the nature of the problem. The more recent models have challenged some of the basic concepts and assumptions around safety. For example, Besnard and Hollnagel (2014) argue that some of the terms commonly used in safety can be dangerously misleading and describe them as myths. They examined six of these: human error, procedure compliance, protection and safety, mishaps and root causes, accident investigation and safety first. Human behaviour they say is conditioned by ever- changing conditions so that performance adjustments can be judged inaccurately if not taken in

context. So, for example, complying with procedures may not always be the safest course to take. Similarly, safety controls built into technologies used may not work in the same way with all people or under all conditions. Also, as accident investigation involves a social process, the causes are constructed within the perspectives of the investigator(s). Moreover, the concept of ‘safety first’ is misleading as the level of safety achieved is always limited to what is affordable.

Being alert to these possible myths is instructive and particularly relevant to a study of safety management in the heavy vehicle transport sector as it is, by its nature, an industry involving a number and range of interactions in a less controlled work environment (compared with a factory or office). As the road environment itself is ever-changing truck drivers must be able to continually to adapt their behaviour to meet safety and efficiency demands.

A range of other theoretical models has been used in injury prevention fields. A recent review of these models, specifically for the purpose of finding models to improve road safety, found that the 121 models that might be applied to road safety can be grouped into seven categories – component, sequence, intervention, mathematical, process, safety management, and system models (Hughes et al., 2015). Each type of model applicable to injury prevention may be particularly suited for differing purposes and contexts. It may not be useful, therefore, to find or develop the best model for injury prevention across the board, but it is worth examining the way that road and transport safety research and interventions have been shaped by theoretical models to see if there are other theoretical constructs that can guide an improved examination of road safety problems and solutions. Hughes et al concluded that none of the systems approach models have been generally applied in road safety even though they show promise for improving safety in this context.

There is increasing support for the idea borrowed from WHS that a systems approach to analysing road injury factors may help to explain the dynamic interaction of the risk factors within a work or driving process and the fact that the conditions built within the system can create risks that sometimes long pre- date the proximal unsafe act that finally resulted in an injury crash, e.g. inadequate brake maintenance, driver payment systems, etc. (Mooren et al., 2009b; Williamson and Friswell, 2013). Other aspects of the WHS systems approach, may also be usefully applied in road safety. The concept of safety culture has been highly influential in WHS on the basis that a good safety culture can enhance or reduce the effectiveness of attempts to address proximal causal factors. In fact, Zohar (1980) argued that effective safety management requires the underpinning of the work process with a management commitment to safety. Beyond this, a climate where workers have a unified set of cognitions about the safety aspects of their organisation is needed to manage safety.

There is a growing body of research on the effects of safety climate and safety culture. Zohar (1980) was among the earliest researchers to demonstrate that employees’ perspectives on the importance placed on safety by managers and how relevant these attitudes were in relation to work performed can have a bearing on safety outcomes. While safety culture and safety climate are terms that some

45 have been used interchangeably, culture refers to belief systems and attitudes, climate can be measured in terms of perceptions of manifestations of safety commitment. Since Zohar’s study of Israeli companies, many attempts have been made to develop an agreed scale for measuring safety climate (Arboleda et al., 2003; Brown and Holmes, 1986; Choudhry et al., 2007; Cigularov et al., 2010; Coyle et al., 1995; Flin et al., 2000; Fogarty and Shaw, 2010; Frazier et al., 2013; Olsen, 2010; Seo et al., 2004; Silva et al., 2004). A comprehensive synthesis of the transport related safety culture literature combined with empirical studies undertaken in 2007 concluded that the following concepts were important aspects of safety culture (Short et al., 2007, p. 1).

Culture and safety have a clear connection. Safety culture is best defined and indexed by an organization’s norms, attitudes, values, and beliefs regarding safety. Effective top to bottom safety communication and interactions enhance safety culture. Terms such as “accident” and “mishap” are often replaced with the terms “crash,” “wreck,” and other more appropriate, straightforward terms in many safe cultures. In many instances, organizations, organizational subgroups, and professions may each have identifiable safety culture. Recognition and certain rewards systems for safe behavior are an effective component of safety culture. Driver experience enhances a safety culture, especially if that experience is with one carrier. Driver retention problems, however, have the potential for degrading a safety culture. Many levels of communicating safety culture are necessary in “remote workforce” industries such as truck and bus operations. Policies, procedures, employee safety responsibilities, and safety messages must be clear and simple.

Safety culture has begun to be defined and examined in road safety research, including in the heavy vehicle transport sector, but there is little evidence of effective practical application of safety culture improvement processes in this industry.

This brief review of workplace safety management has attempted to explore the models and concepts that have been thus far developed, largely in workplace safety management. Accident models that help to explain how and why accidents occur have undergone a number of iterations since Heinrich developed the Domino Theory in 1931 (Heinrich, 1931). These models have increasingly recognised the need to appreciate the dynamic and complex interactions that characterise many modern day work processes in order to determine how to respond to risks in these processes. Thinking in terms of whole system processes, rather than just analysing the components of the system itself has generally not been applied in either work-related or public road safety. Finally the safety culture or climate of organisations has been

emerging as an important aspect of safety management, overlaying work processes and influencing safety outcomes and the effectiveness of other safety management measures.

2.4.2 Identifying weaknesses in organisational safety defences through accident analysis The models describing the causes and circumstances of accidents are important because how safety is conceptualised will influence how we investigate and analyse accidents. Work health and safety professionals widely believe that by conducting investigations into all relevant factors present leading up to the time of an injury event, organisations can identify systemic and individual factors involved in injury incidents. How these investigations are designed will influence what is found.

Whether work systems are simple or complex, Kjellen (1984) argued that it is useful to understand safety risks as deviations from the normal work process. That is, where the work process is a planned set of actions to achieve a planned result, a safety risk represents an unplanned variation to the process. Predicting the kinds of deviations that could occur would be the key to enable the prevention of accidents. Hale et al (1997) incorporated this idea into a framework for safety management in response to the growing interest in auditable safety management systems in the mid-1990s, particularly in Europe. They recognised that the research literature on safety management was fragmented and needed to be assembled into some order. Responding to this need, and accepting the notion that safety risks are deviations from the normal work process, the authors developed a model that could guide the way accidents are analysed so that effective processes to address hazards could be more readily identified. Figure 2.6 illustrates the model.

Figure 2.7 Deviation model (Hale et al., 1997, p. 128)

47 The Deviation Model defines a work process where embedded hazards continually present risks as deviations to the normal process. It provides the places in the process to look for hazards or for factors contributing to accidents. The actions to eliminate, control or avoid hazards and risk factors is an ongoing stabilisation process in which recovery and learning results in redesigning of the system, then managing each subsequent deviation.

Hollnagel (2002) examined various methods of accident investigation and analysis in terms of how safety professionals would choose a method, or methods, suitable for their particular work processes. He identified three types or models of accident analysis. The sequential models are those that examine linear chains of events or construct event trees to understand causal factors and the links between them. These models are used to trace back from proximal (often human) accident factors to processes and conditions that enabled the proximal factor to lead to the event. These models are the simplest methods of investigating accident causation factors as described in Reason’s accident trajectory model, identifying the holes in the safety defence system. However, they do not cater well for incidents that involve complex non-linear risk factors.

Hollnagel’s epidemiological models examine carriers or barriers (of risk or disease) and latent conditions to identify safeguards to be put in place or strengthened. These models tend to treat the ultimate “unsafe act” as a result of latent conditions rather than an accident causation factor. In other words the work process and conditions shape and predict the final action in an accident event. The emphasis is to prepare or repair the unsafe conditions in the work process or environment. This model is not limited to linear chains of causation factors, but it lacks the dynamic attributes that the third class of models exhibit.

Hollnagel’s systemic models focus on looking for variations of processes and conditions (compared with normal operations) associated with accidents that continually need to be monitored, controlled or eliminated. This model assumes that humans behave in variable ways and that an understanding of how and why they behave can guide the development and maintenance of systemic safety measures. In other words, the system should be designed to suit the way humans interact with it. These models also call for continual examination of the interaction between human users and system performance.

Accident investigation methods are shaped by the theoretical model adopted by the investigators. Understanding all factors leading to a harmful event informs the safety improvement process. The objective is to find all systemic weaknesses in the work process that may provide the conditions under which a triggering factor can result in an accident. Sklet (2004) reviewed fourteen different methods for this kind of incident investigation.

Some models, such as the Tripod Beta method (illustrated in Figure 2.8), are based on the idea that all accidents are caused by organisational failings. This method is an example of a sequential model.

Figure 2.8 TRIPOD Beta accident causation model (Reason et al., 1989)

Most of the models examined by Sklet focus on safety barrier failures within the organisation. Moreover, Skelt concluded that “major accidents almost never result from one single cause, but most accidents involve multiple, interrelated causal factors.” For this reason, a root cause analysis has been criticised because it is unlikely that an accident could be traced back the chain to discover a single root cause (Sheridan, 2008). That is, although chains of causality can be identified, the goal of identifying a single root cause is questioned because finding a single deviation or latent pathogen may not eliminate all sources of risk. Even though the linear style of root cause analysis can find multiple sources of system failures, Hollnagel (2008) argues that setting up barriers to known risks is reactive, and while it would still need to be done, needs to be complemented by proactive safety management.

Hollnagel and others have argued that resilience can be built into a safety management system to anticipate risk and make adjustments and quick recovery without upsetting the dynamic production processes. Resilience engineering is a paradigm that treats safety as a core value, rather than a quantitative measure, although they are not necessarily mutually exclusive. Hollnagel says, “safety is a chronic value under our feet that infuses all aspects of practice (2006, p.5).” This approach is one of continual assessment and adaptation to ever changing processes, examining interactions across factors rather than treating factors as separable and independent. However, there is yet no evidence of the efficacy of any resilience-engineering model.

2.5 What is a safety management system?

The field of workplace health and safety has embraced implementation of a safety management system (SMS) as a productive approach to reducing workplace incidents. The International Labour Office (ILO) (2001), defines SMS as a set of interrelated or interacting elements to establish safety policy and objectives, and to achieve those objectives. An SMS typically includes policies, strategies, practices, procedures, roles and functions associated with safety (Kirwan, 1998). The primary aim is to reduce accidents, injury and exposure to risk. Hale et al. (1997, p. 121) say that “safety management is seen as a set of problem solving activities at different levels of abstraction in all phases of the system life cycle.” Safety management can be understood as actions to meet safety objectives through processes and procedures. In essence, a safety management system is a set of processes and procedures that aim to control safety threats.

49 There is increasing belief that promoting a good SMS across the whole system of work may be key to reducing incidents and injury as this has been introduced in some industries such as mining, aviation, rail and oil and gas (Grote et al., 2009; Helmreich, 2000; Hudson, 2007; Joy, 2004; McDonald et al., 2000; Mearns et al., 2003). However, evidence to support any specific comprehensive set of elements of safety management is still fragmented (Robson et al., 2007). For example, Bottani et al (2009) compared companies that had implemented an SMS program to address their WHS problems with those that had not, and showed that SMS companies had better performance on a range of outcomes including the number of accidents. However, the specific elements of the SMS were not well defined. Other studies showed that safety outcomes vary with the degree of safety commitment by managers (Mearns et al., 2003; Newnam et al., 2008; Seo et al., 2004). Fernandez-Muniz et al (2009) in a WHS study of 455 Spanish companies showed that a good safety management system, and safety practices permeating and enhancing safety culture, improve safety performance and reduce accident risk. The findings of these and other empirical studies will be discussed more fully in Chapter 3.

Overwhelmingly, the models describing SMSs have come to conclude that taking an SMS approach, dealing with hazards, safety risk and accidents requires consideration of the system of work management and socio-technical environment affecting individual risk factors, as discussed in Section 2.4. Roland and Moriarty (1990) emphasise that the system safety involves a proactive application of technical and managerial skills to anticipate and control hazards throughout the life cycle of a project, program or activity. The emphasis is on the control of hazards from the very beginning or design phases of the system (any system) and for as long as the system operates.

Hale et al (1997, p.125) argued that a sound safety management system should meet the following criteria.

1. It should model the complex, dynamic systems that the SMS exists in. 2. It must be able to focus in on elements of the system without losing the links to the whole model. 3. It must provide a common language to describe and model all aspects of the system. 4. It must be compatible with existing ideas and principles in safety management, quality management and the idea of the learning organization. 5. It must provide links to standard concepts used in management texts, e.g. the distinction between policy, procedures and instructions. 6. It should model both the primary (technological) processes with their risks and the management decisions which control them. In summary, a review of the SMS concept by Hale et al., as applied in workplaces showed the important aspects of a good SMS and the benefits that these programs, and while yet untested, can potentially deliver. An important finding of this review is that it is important to develop safety management systems that reflect and effectively address the way in which safety hazards manifest and emerge in operation of the whole system.

2.6 Applicability of SMS to the road freight transport sector

Hazards in the transport process are often complex factors that form interdependent process chains as discussed in Chapter 1. An example of this is a truck fatality involving driver error due to fatigue – which in turn could stem from lack of rest due to irregular work hours, heavy workloads, poor scheduling, monotonous roads, difficult or rough vehicle to drive, little opportunity for rest and recuperation between work shifts and pay systems that encourage driving for long hours.

The complexity of influences on safety outcomes is exemplified in an analysis by Williamson (2007), which showed that Australian heavy truck drivers were more likely to use stimulants while driving if they had difficulty managing fatigue, and if they were paid on an incentive basis that encouraged them to do more work. Such examples demonstrate that a number of interrelated risk factors occur at different points and levels in the transport process and that there are a number of possible controls that can mitigate the hazards occurring at each of these points and levels. Following this argument, the incidence of driver fatigue is not likely to be overcome by better scheduling of work and rest time alone, because if the incentive based payment system is left in place, the risks are likely to persist. A safety management system that helps to continually identify interdependencies between risk factors and to adapt to work conditions and processes should be the ideal.

A safety management system suitable for heavy vehicle transport operations should be based on the idea that any WHS incident results from a set of interacting causes, not simply from a proximal error occurring at the time and place of the incident (Mooren et al., 2014; Newnam and Goode, 2015). For example, if a heavy vehicle driver is injured in a crash, that outcome will reflect the road and traffic environment at the time but also, perhaps, the safety features of the vehicle and in turn the vehicle-purchasing practices of the organisation and potentially its size and profitability, as well as the driver’s skill and professional attitude. Additionally, factors include staff and driver selection and training practices, the company’s prioritisation of safety, as reflected in work scheduling, incentives for drivers to work unsafely, the communication of safety information to drivers, as well as a range of other factors not directly linked to the road incident. That is, although a truck driver on the road may be remote from a company site, the driver’s safety can still be influenced by a range of management and work practices.

Based on the experience of other transport sectors, such as rail and aviation, as well as in other industries it can be concluded that a systematic approach should be taken in the road freight transport sector to manage these factors to reduce the risk of incidents that result in injury. These other transport sectors have adopted the SMS approach. For example, all accredited airlines and railways in Australia are required to have safety management systems. The prescribed SMSs have not been formally evaluated for safety outcome performance but have showed some promise for influencing attitudes towards safety. For example, a small regional airline in Australia implemented an SMS initiative in their Sydney based operations and not in their Melbourne based operations and had an auditor compare the differences in

51 staff perceptions of safety management. Criteria for the evaluation included safety culture, risk perception of safety hazards, willingness of staff to report safety hazards, action taken on identified safety hazards and staff comments about safety management within the airline. The evaluation showed that there was a much more positive perception of the airline’s safety management in the Sydney based operations (Edkins, 1998). To be consistent with its membership obligations to the International Civil Aviation Organisation16 and informed by this evaluation study, the Civil Aviation Safety Authority (CASA) of Australia developed a safety program outlining an SMS that all licensed airlines must implement (CASA, 2011).

Also the Office of the National Rail Safety Regulator (ONRSR) provides a guideline on the preparation of SMSs to explain how to comply with rail safety regulations and how to develop safety management systems (ONRSR, 2013). As the regulator requires root cause analyses of incidents to attempt to identify systemic weaknesses in injury prevention defences, an additional resource is incorporated to guide the investigation and root cause analysis of rail safety incidents. This manual of “contributing factors” to safety incidents in the rail industry aims to strengthen systemic safety management defences against risk. As all Australian rail operators must be accredited to operate, these are considered to be practical tools for assuring compliance with safety management requirements. But no formal evaluation of this initiative has been undertaken to demonstrate that it has reduced rail safety incidents.

Both the aviation and rail SMSs emphasise the importance of building a strong safety culture. This is consistent with the body of safety culture research discussed in Section 2.4.1 that suggests that safety culture is an important success factor in safety management.

A Joint Australian/New Zealand Standard, AS/NZ 4801:2001 for Safety Management Systems (SMS) was prepared by Joint Technical Committee SF-001, Occupational Health and Safety Management. The objective of this Standard was to set auditable criteria for an workplace health and safety management system (BSI, 2014). Accreditation to the Standard is voluntary. There is no evidence of the Standard having been formally evaluated and, by its own description, it does not ensure compliance with all legal and other obligations. So, while applying this Standard may improve safety management practices in organisations, it is yet to be demonstrated.

The developments in safety management in other transport industries, such as aviation and rail transport, are worthy of consideration for applicability in the road freight transport industry. However, many of these developments of SMS’s and accreditation standards have yet to be empirically tested.

2.7 Conclusions

As illustrated in Chapter 1, the decisions of drivers and managers working in the road freight industry are influenced by a range of underlying characteristics of the work systems and the nature of the industry. This

16 The ICAO safety management manual can be obtained at http://www.icao.int/safety/SafetyManagement/Documents/Doc.9859.3rd%20Edition.alltext.en.pdf

industry sector is fraught with inherent safety risks. Regulatory and other efforts to contain or reduce the high level of trauma associated with this industry, such as industry codes and accreditation schemes, have been made by governments and industry associations. However, there are still high levels of injury and fatal incidents.

The risk environment in the trucking industry, like many others, is complex, dynamic and involves decisions affecting safety being made by a chain of players, including transport buyers, consignors, schedulers, contractors, managers and drivers. This means that safety-critical decisions are influenced by people who do not have a direct responsibility for safety. Too often the pressures placed on drivers to work longer hours, drive at unsafe speeds, and engage other unsafe practices, are intensified by unrealistic work demands.

There is also little acknowledgment that heavy vehicle driving occurs in an workplace context, with most focus on truck driver safety management being on-road (typically the domain of road authorities). The heavy vehicle road transport industry has several unusual characteristics that are likely to affect safety management for drivers such as the tendency towards small companies, complex layers of subcontracting, remote work without direction supervision, and irregular and shift work, for example (Adams-Guppy and Guppy, 2003; Mayhew and Quinlan, 2006; Quinlan, 2001). Economic and regulatory factors can strongly influence safety, particularly in the transport industry where market competition can be intense (Australian Transport Safety Bureau, 2008).

The primary responsibility for managing safety rests with managers and drivers. This responsibility is codified in transport and workplace health and safety legislation, but transport safety regulations do not prescribe specifically how to manage safety except on the road, and neither do WHS regulations in relation to managing safety in road freight transport.

A substantial amount of research has been done in the workplace safety field to pave the way for the development of a safety management system. Theoretical modelling literature has advanced to a point of maturity that recognises the multi-dimensional and dynamic nature of work. Models and definitions of safety and risk have increasingly recognised the complex and dynamic processes involved in managing work. WHS thinking, and increasingly the thinking in road safety, acknowledges the role of a broad range of factors, many of which are pre-existing, that can influence the risk of active failures (such as the making of an error or the breakage or malfunction of equipment), or influence the consequences of these failures. The SMS approach also acknowledges that successful management of safety requires consideration of a broad range of factors that could or do play a role in safety, for better or worse.

A growing body of research shows that a safety management system can feature flexibility and resilience to manage expected and unexpected risks. This requires organisations to be continually anticipating risk and be responsive and proactive in eliminating risks. The nature of road transport tasks involves exposure to a relatively uncontrolled work environment – public roads. It is therefore important for truck drivers and

53 managers to continually anticipate risks and adapt the process to avoid injury crashes. As was found in the development of the STAMP model for the aeronautics industry, safety management in trucking is best defined as a continuous adaptive control task.

There is little evidence-base for a comprehensive approach to safety management in the road freight transport industry, or any other industry. However, WHS studies in the trucking industry and other industries provide some grounding in what management and organisational characteristics might influence safety performance.

A number SMSs and accreditation standards are in existence containing a range of management elements but these have not been evaluated. Therefore, there is little evidence even in WHS where most research has been done, about what actually is important for managing safety risk.

Building on the evidence of the promise of SMS to improve safety, the premise of this thesis is that it is possible to develop an effective and suitable SMS that will improve safety outcomes in trucking companies. The study approach is to identify the specific components of a system of safety management that distinguish between good and poor safety performers in the road freight industry, and thus identify evidence for what should be included in a safety management system particularly suited for this industry.

This chapter explored the theoretical basis for how safety and risk are understood in road and workplace contexts. There has been an evolution of safety definitions and models in the literature over the past eight or so decades. Conceiving of safety as the absence of harm has led to the development of models focused on identifying and recognising and reducing sources of harm. Increasingly, models took into account, not just the immediate causes of the injury incident, but risk factors or latent conditions that are found within the system or process that make the immediate cause more likely, or more likely to have an adverse consequence. These systemic factors or conditions may be physical, such as faulty brakes, or non-physical, such as a lack of safety commitment by managers. These factors have been considered as root causes in an accident trajectory, where the combination of those factors and a final trigger (usually a human error) results in a harmful event.

Therefore, each of the risk factors in causal chains has the potential to be addressed with a safety intervention. The further work on these models recognised that chains of risk factors may not always be linear and that any one incident usually involves more than one interacting causes.

The chapter also reviewed the literature particularly relating to theoretical developments in safety management. It found that despite widespread use, there is little research on what elements of an SMS might make an improvement to safety outcomes at all, much less for particular industries or sectors.

Chapter 3: Strategic Scientific Literature Review (Study 1)

This chapter discusses the findings of the first study, a strategic review of the scientific safety management research literature.

3.1 Introduction and aims of this Study 1: systematic review of scientific literature

In order to address the primary focus of this thesis, the development of an evidence-based Safety Management System for heavy vehicle operations, it is necessary to identify the major components of safety management that produce good safety outcomes, i.e. demonstrate reduced crash/injury rates, in the heavy vehicle sector as these may be different from those identified for other industries.

No study has yet developed a transport management system through a scientific research process, nor have any transport Safety Management Systems been scientifically evaluated with quantitative methods. Consequently, the outcomes from the current work will advance the knowledge base in relation to road safety and heavy vehicle crashes and fleet management practices and policy.

One of the problems with advancing the SMS approach to safety improvement, is that any number of plausible links between safety and work or management practices might be proposed for inclusion within the coverage of an SMS, but there was little empirical guidance about which are most important to manage safety in reducing real world crash outcomes. This lack of evidence had been identified in the field of workplace safety generally (Hale et al., 1997; Robson et al., 2007), but also with regard to the broad category of work-related driving (Grayson and Helman, 2011). For the heavy vehicle road transport industry, where SMS have hardly been applied, this was particularly so and was a concern because of the high crash and injury risk associated with heavy vehicle drivers’ work.

The starting point for identifying the safety management components that are the most effective (crash-reducing) ones to include in a safety management system for heavy vehicle transport operations was to examine prior empirical research evidence. Thus, the aim of this study, Study 1, was to examine existing workplace safety literature to identify specific characteristics that have been demonstrated to result in good safety outcomes across workplace settings. The results of this review will then be used to form the basis of the safety management framework evaluation used in a survey of road transport operators (Study 2) to identify the safety management characteristics that distinguish between good and poor safety performing companies. The literature review was a critical aspect of the research as it informed the specific measures used in the comparison of higher and lower safety performance companies.

The findings provided in this Chapter is the evidence found in the scientific research literature that identified safety management items that have made demonstrable differences in safety outcomes. This review was conducted in two stages. The first stage looked at published studies up to the end of

55 2011. These results were published in Mooren et al (2014). The second stage updated the review to include more recent literature published between 2012 and 2015 inclusive.

3.2 Literature review methods

An extensive search of the scientific literature as well as industry publications (grey literature) was undertaken. This included electronic and physical library research. A total of 6,264 references were found by searching Google Scholar and the University of New South Wales Library online service, Sirius, using the key words, “safety management system, safety culture, risk management, occupational work driving, study results, significant finding.” Sirius enabled a simultaneous search of databases including: Scopus, ScienceDirect, Proquest Research Library, APAFT, Business Source Premier, Social Sciences Citation Index, Web of Science, JSTOR and Primo Central Index.

The initial broad search included studies relating to patient safety, environmental safety, chemical and nuclear safety, financial risk and consumer safety. Based on titles, the articles that did not pertain to work-related safety management, or duplicates were removed. In addition, those that examined road safety issues in a general or public health context were excluded. Only articles that were published in English were selected. A total of 532 references remained after this cull.

The abstracts of these were reviewed in terms of whether or not they met the criteria for closer review. In the first instance the criteria included the identification of specific measures that might predict safety outcomes. Secondly, an assessment of transferability to safety management in the heavy vehicle transport sector was made. Finally, those studies that developed but did not empirically test safety climate scales or provide any evidence-based outcomes from evaluations were excluded.

In addition, the original scientific study literature that was included in relevant systematic reviews (e.g. Flin et al. (2000), Grayson & Helman (2011), Guldenmund (2000) and Murray et al. (2009b)) was obtained for first-hand examination. Any new articles found therein were also subjected to the same criteria for inclusion.

After a review of the abstracts, in 2011, for the papers that met the empirical/evidence-based outcomes criteria (original research) a total of 124 papers were selected. These were categorised as those that examined observable management systems or characteristics (SMS) and those that studied subjective characteristics (safety climate). There were 81 papers classed as SMS studies and 43 classed as empirical safety climate studies. This initial list of references were then provided to research review team that confirmed the thesis author’s identification of those papers that have statistically significant (i.e. with p-values of less than 0.05) predictors of injury outcomes were classified either as organisational characteristics or specific interventions. Successive initial meetings with the study review team (that included the thesis supervisors) identified the final list of the studies to be included in the review. A total of 42 references formed the basis of this initial literature review.

Because this project has continued over a number of years (2009-2016) the initial review, reported in Mooren et al (2014) was updated by the thesis author to include more recent studies, using the same search criteria and specifying inclusion of items published from the beginning of 2012 onwards. Applying the same selection criteria as in the initial review, the new search yielded an additional 26 studies that are relevant to the project. The findings of these studies were integrated with the original 42 studies. Therefore, this Chapter analyses the findings of 68 studies related to safety management.

Analysis of each study involved identifying the characteristics found to be statistically significantly associated with better safety outcomes, i.e. reduced injury outcomes. The number of times a particular characteristic was identified across studies was counted. Where this process identified a particular characteristic in two or more studies it was judged to be sufficient evidence to be considered for inclusion in the survey in Study 2. Where a study looked at the effects of comprehensive SMS programs, individual elements were identified and included to avoid double counting and because it is possible that characteristics interact in more comprehensive SMS programs.

3.3 Results of the strategic literature review (Study 1)

Analysis of the 68 original research articles resulting from the review process showed that they fell into three distinct groupings. The first grouping contained studies in which the organisation was the unit of analysis. That is, they looked at the differences in characteristics between companies and the measure of outcome was company safety performance. The second grouping included studies in which drivers or workers were the unit of analysis and involved individual surveys about their attitudes, behaviour and/or, experiences and measured the effects on individual employees. The third grouping involved evaluations of the relationships between specific safety management-related interventions and safety outcomes. The methods in these studies were mostly pre-post data collection and analysis of changes in intervention and control groups.

These three groups of studies were examined separately as they each provided separate insights into safety management. The first group of studies provides the strongest and most direct evidence of relationships between different characteristics of companies and safety outcomes at a company level. The second group of studies show the safety effects on individual workers of particular safety characteristics. The third group were evaluation studies that show whether or not particular safety- related interventions changed safety outcomes.

Taken together, the results of the three of study groups provided converging evidence on the factors that are most likely to be important in influencing safety outcomes. The results of each of the three groups of studies are summarised in Tables 3.1 - 3.3. Each table includes a list of the studies reviewed including the methods of study, the participants, the items studied, the outcomes and the predictors found to be associated with positive safety outcomes that had statistically significant

57 results (at a p-value greater than 0.05). The results were then summarised to compare the evidence found for each individual safety management characteristic in each type of study as shown in Table 3.4.

Finally, the results found for the individual safety management characteristics specifically in the heavy vehicle transport industry were also summarised. These are presented in Table 3.6.

3.3.1 Studies of organisational characteristics and safety results at organisational level A summary of 25 studies relating to the relationships between organisational safety efforts and the likelihood of predicting injury outcomes is presented in Table 3.1.

Out of the 25 studies summarised in Table 3.1, sixteen (16) or 64% were cross-sectional surveys involving self-reported safety outcomes. Five (5) studies (20%) used existing crash and other safety related data for analysis. One (1) study (4%) involved a longitudinal study of accident rates, and one (1) study (4%) used longitudinal crash data as well as a cross sectional survey. One (2) study (8%) used a combination of audit data and cross-sectional survey methods.

Twelve (12), or (48%) of the studies focused on heavy vehicle transport safety and two (2) studies (8%) were about light vehicle fleet safety. The remaining eleven (11) studies (44%) were conducted with other industry sectors such as manufacturers, hospitals and chemical companies.

The types of outcome measures varied between the studies. These can be grouped into three categories. A total of fifteen studies (15) out of 25 (56%) used accident and injury outcome measures (including self-report incidents, n=5). Five (5) studies (20%) used behavioural change indicators, and five (5) studies (20%) used perceived measures of safety culture. A total of twelve (12) studies (48%) involved the heavy vehicle transport industry.

Extraction of the significant findings17 from the 25 studies indicates the most likely predictors of safety outcomes at an organisational level include:

management commitment/leadership/safety climate (8 studies, 32%), employee participation (8 studies, 32%), risk and incident analysis and corrective actions (7 studies, 28%), incentives, rewards (6 studies, 24%), safety management system (5 studies, 20%), vehicle/work environment conditions (4 studies, 16%), the size of the organisation and freight type (3 studies, 12%) safety management accreditation (3 studies, 12%), policy enforcement/discipline (3 studies, 12%), remuneration/financial performance (3 studies, 12%),

17 Note that some of the studies included a number of possible predictors.

safety scheduling and planning (2 studies, 8%) training (2 studies), and hiring and retention practices (2 studies).

In summary, the two most studied safety management characteristics in this group were management commitment/leadership/safety climate and employee participation (in WHS).

59 Table 3.1 Studies of safety factors and outcomes with company as the unit of analysis

Study Study design Participants Measured Outcome measures Significant predictors associated with good safe~ outcomes (Abad et al., Longitudinal study comparing 149 Spanish businesses OHSAS 18001 certification Accident rates/employees OHSAS 18001 certification• 2013) companies that have been certified from construction. Proportion of fatal (of accident rates) to OHSAS 18001 and those that manufacturing and accidents haven't for differences in safety service sectors Average number of lost performance and productivity using workdays linked insurance and standards Sales/employees certification records (productivity) (AI-Refaie, Cross sectional surveys: 176 surveys of managers Safety activities, safety management system Safety self-efficacy (belief Management commitment affects self -efficacy via 2013) demographic and rating of safety and workers in medium (SMS), reward system, safety reporting about one's own safety safety activities• and safety awareness and safety culture items using a Likert scale 1 sized and 148 surveys of system, blaming culture, management abilities), safety behaviour behaviour through teamwork and supervision• (strongly disagree) to 5 (strongly managers and workers in commitment, continuous improvement, (risk behaviour and Interrelationships between team members affects agree) on the impact of safety 24 large sized Jordanian hannonious interpersonal relationships, compliance to rules), safety self-efficacy via safety activities• and safety culture factors on other factors and companies employee empowerment, supervision and awareness (risk perception) awareness through teamwork• Continual outcomes using a structural teamwork improvement affects safety awareness via teamwork* equation modelling method. and safety behaviour through teamwork and SMS* (Arboleda et Cross-sectional survey of influences Drivers, dispatchers and Safety (fatigue) training, driver scheduling Perceptions of safety Safety (fatigue) training.. • , opportunity for safety al., 2003) on individuals' perception of safety safety directors of autonomy, opportunity for safety input and culture input•. top management commitment•.. culture within organisations transport companies top management commitment n=116 us Baas and Comparison of crash rates of National Heavy Vehicle Accreditation Crash rates/truck year NHVAS accreditation••a Taramoeroa accredited and non-accredited (n=8,178 crashes) Scheme accredited companies (mass, (2008! o~rators (Australia! maintenance or both! Banks (2008) Audits of organisational practices, Public and private sector Policies, hiring criteria, induction, training Self-reported errors, fatigue Companies having comprehensive risk management interviews and survey of organisations. (n=4) needs assessment, incentives/disincentives, and violations strategies • employees. A triangulation of the Employees (n=679) journey planning, vehicle selection & data sets was used to assess fleet (Australia) maintenance, and incident analysis safe ractices Bruning Examination of US truck crash data Freight truck crashes with Company size, vehicle types and conditions, Crash rates and driver (Specialised) Carrier size.. . defective/ old equipment (1989) on and company characteristics. >$2,000 damage (n= 540 and financial performance related, equipment related, for small and medium sized carriers•.. and driver general) (n= 864 weather related and tenure.. specialised) (US) managerial related crash factors Corsi et al. Data analysis of the relationship Trucking companies Financial performance, driver wages Satisfactory US Safety Financial performance.. (2002) between US motor carriers' safety carrying general and Compliance Reviews Driver wages.. performance and financial specialised freight ~rformance (n=700! (US! Crumand Cross-sectional survey to validate a Trucking companies Regular driving time, trip control, quality of Self-reported dose calls, Starting the workweek tired*, inadequate rest area on Morrow truck driver fatigue model (of risk) (n=37 4 safety directors & rest, scheduling demands of commerce, perceptions of fatigue journey route .. and size of delivery window.. (2002) 279 drivers) (US) driver economic or personal factors, carrier economic factors

60 Study Study design Participants Measured Outcome measures Significant predictors associated with good safety outcomes (Corsi et al., Regression analysis of US SafeStat Data 157,292 US motor Effects of unionisation DRSEA scores Union membership has a negative impact on 2012) against driver safety evaluation area carriers (78 VHSEA scores DR SEA and VHSEA scores* (meaning unionised scores (DR SEA) and vehicle safety unionised) Crash rates carriers have better safety performance) evaluation scores (VHSEA), and BASIC Unionised motor carriers have a lower crash rate* scores (as safety performance indicators) de Pont Analysis of Tasmanian truck crash data to Truck crashes for Defects and mechanical failures, freight Truck crashes between the Defects and mechanical failures, freight type b (2005) determine if vehicle types & years 2002 & 2005 (n type years 2002-2005. configurations influence crash risk. = 162) (Australia) (Fernandez- Cross-sectional survey to test perceptions Safety managers of SMS including: policies, incentives, safety Employee satisfaction SMS** Muniz et al., of the relationship between selected companies in training, communications, preventative with: the number of personal 2007a) safety management elements services, building and planning, emergency planning, injuries; material damage; industrial sectors enforcement, and incident reporting employees' motivation; and (n=455) (Spain) absenteeism or lost time (Fernandez- Cross-sectional survey of safety Safety managers of Leadership, pro-active risk management. Proactive risk management, Leadership to proactive risk management• Muiiiz et al., managers to measure the effects of organisations in safety participation, safety compliance, safety participation, safety Proactive risk management to safety participation* 2014) leadership and safety participation on various industries compliance, (self-reported) Participation to safety compliance* safety outcomes (n=188) (Spain) safety outcomes, employee Safety compliance to safety outcomes* satisfaction Safety outcomes to employee satisfaction* Geldartet Cross-sectional survey and analysis of Manufacturing Monitoring injury statistics, auditing, safety Objective data: lost time Monitoring injury statistics**, auditing**, safety al. (2010) organisational approaches in that companies (n=312) awards, employee surveys, worker injury frequency rates awards**, worker participation in WHS** characterise safer manufacturing (Canada) participation in WHS worl

61 Study Study design Participants Measured Outcome messures Significant predictors 888ociated with good safety outcomes Moses and Cross-sectional study, using US Safety US motor carrier safety Firm size, firm age, for-hire/not freight type, Crash rates per truck-mile Firm size•. for-hirefnot freight type•, filing crash reports•. Savage Review audits of trucking companies to audit records filing formal crash reports, disciplining drivers disciplining drivers• (1 994) analyse effects of firm characteristics (n=75,577)(US) for preventable crashes on crashes Naveh and Longitudinal and cross-sectional ISO 9000 certified, and ISO 9000 certification. return on asset (ROA) Fatal, injury and towaway Certification• and higher ROAs• Marcus studies about whether ISO 9000 quality no~rtified trucking crash rates; improved (2007) management certification effected companies (n=840) (US) financial performance crash rates (Peignier et Cross sectional survey identifying and 21 1 Canadian HAZMAT Carrier size - small and large numbers of Use oftechnologies Large companies more often: al .. 2011) comparing safety management carrier companies 54% employees Written procedures Have a means of communication with the driver.. practices of hazardous materials having< 10 employees, Record keeping Record speed and deceleration.. • transport companies and 44% ~ 10 employees Procedures Use vehicle maintenance software .. and 2% unknown size Use of regulatory information Keep incident register.. Emergency preparedness Investigation following an incident• Written procedure for subcontractor selection.. Information from professional association and government•.. Information from wori< safety committee.. Member of professional association.. • Emergency response team• Use simulation training for emergency response" Saksvik et Cross-sectional survey to measure Managers of various WHS activities. motivation and training, Self-reported occupational The use of SMS assessment•.. al. (2003) WHS activities, motivation and training, organisations (n=1,789) cooperative implementation, implementation injuries and sickness cooperative implementation, (Norway) status (SMS assessment) absences implementation status. and WHS outcomes Silva and Cross-sectional survey and Employees at 4 chemical Safety as a value. perception of organisational Self-reported accident rates Actions matching espoused values (safety climate)... and use Lima (2005) comparative case study comparing companies (n = 328 safety practices, learning strategies from and employee perceptions of accident data for prevention purposes•.. espoused company values with participants) (Portugal) accidents employees' perceptions of practices and accident rates Vredenburgh Cross-sectional study to determine Hospitals (n=62) (US) Wori

62 Study Study design Participants Measured Outcome measures Significant predictors 888ociated with good sat outcomes (Wachter and Cross-sectional studies testing the Study 1: 342 safety Employee involvement/influence Self-reported accidents (total All measures in study 1 significantly" predicted Yorio, 2014b) relationship of safety management managers (various sectors) Pre & post-task safety reviews recordable cases) accidents. with the majority significant.. • systems and practices (1 0) to safety Study 2: 144 supervisors Safe work procedures Accident severity (days away All measures in study 2 significantly* predicted outcomes. Study one surveyed safety from 29 companies and Hiring for safety restricted or transferred) accidents. with the majority significant •• managers on perceptions and actual 650 employees (US) Cooperation facilitation Worker engagement mediates the effects of safety accident data. Study two surveyed (various sectors) Safety training management practices.. supervisors and employees on Communication/info sharing perceptions and self-reported Accident investigation accidents. Detection and monitoring Safe task assi nment Wills et al, (2005) Cross-sectional survey to examine the Employees in 3 Communication and procedures; work Safety climate Management commitment*, communication and factor structure of an existing survey organisations (n=329) pressure; management commitment; support.. measure of safety climate (Australia) relationships, driver training; rules Zohar ( 1980) Cross-sectional survey to test the Employees (400) of 20 Perceptions of: safety training, work Safety rankings by 4 expert Perceived safety (training) relevance to the job •••. efficacy of a 40-item safety climate companies (various pace, status of safety committee & safety inspectors management commitment*.. measurement tool. industries) (n=20) (Israel) officer. staff promotions. level of risk, management attitudes, social status (in workplace) Reported p-values: •p 0 < 0.05. ..P D< 0.02 ... P 0 < 0.002 • While p-values for the employee survey were <0.05, this study triangulated data from the survey, interviews and audits to produce these findings. •While tests for significance were not reported. the estimated savings was based on a robust model.

63 3.3.2 Studies of organisational or personal characteristics and safety results at employee level A total of 34 studies were found that focussed on individuals and groups of workers or managers rather than effects on organisations. These results are described in Table 3.2. While the findings from these studies do not address characteristics and outcomes that are associated with organisational differences they do identify individual characteristics that signify areas that could be productively addressed in efforts to improve the safety performance of organisations.

All but three studies were analyses of existing data using cross-sectional methods. The participants of thirteen (13) of the studies (38%) were drivers of heavy transport vehicles or heavy truck crashes. One (1) study (3%) involved light truck drivers. Seventeen (17) studies (50%) were conducted with workers in various or multiple industries, including oil and gas, grocery retail, government, metal processing and grain industries. The remaining three (3) studies (9%) focussed on light vehicle work- related driving safety.

Nine (9) studies (26%) used outcome measures relating to self-reported crashes or injuries. Six (6) studies (18%) used actual recorded injury or accident outcomes data. Twenty-three (23) studies (68%) used outcome measures involving self-report of other safety behaviour-related outcomes. Three (3) studies (9%) used safety perception as an outcome measure. (Note that a number of studies used more than one outcome measure.)

The significant findings from the 34 studies were extracted and compared. Extraction of the significant findings from the 34 studies indicates the most likely predictors of safety outcomes at an individual level include characteristics relating to the following:

management commitment and attitudes to safety, leadership and trust (14 studies, 41%), safety climate, group norms or worker attitudes (11 studies, 32%), scheduling and work demands (6 studies, 18%), remuneration (5 studies, 15%), vehicle conditions (4 studies, 12%), age or experience of driver (4 studies, 12%), physical and psychosocial work environment (4 studies, 12%), safety policies and rules (4 studies, 12%), employee involvement (4 studies, 12%), safety communication (3 studies, 9%), crash and violation/licence history (3 studies, 9%), company size/freight type (3 studies, 9%), safety training (2 studies, 6%), return to work policies (1 study, 3%), and

perception of a safety management system (1 study, 3%).

In summary, studies at the individual worker level of inquiry found that the research into the safety influences on workers and drivers is largely focused on safety culture and safety climate factors. The most predominantly studied characteristics in this group were about safety culture/climate, including management commitment, leadership, trust, attitudes, or perception of a safety management system.

65 Table 3.2 Studies of safety factors and outcomes with the individual as the unit of analysis Study Study design Participants Measured Outcome measures Significant predictors associated with safety outcomes Belzer et al. Cross-sectional survey of drivers and examination of firm Drivers in one Age, gender. race, marital status. base Crashes Higher rates of driver pay••. pay (2002) data to determine the relationship between crash trucking company pay, pay increase from period 1 to increases.. . and driver tenure.. occurrence and driver characteristics, employment observed for two 13· period 2, miles driven per month, history, and driving activity. month periods dispatches per month, Winter. hiring (n=11.540 =92.528 dale, tenure with firm, prior moving person-months) (US) violations (only for a subset of the data), driving experience prior to hire (only for a subset of the data) Bjerkan (2010 ) Cross-sectional survey testing the perception of 152 Offshore oil & gas Physical work environment, safety Self-reported health and injury Perceptions of physical work safety climate items workers (n=9,945) climate. psychosocial work outcomes environment'.. . psychosocial work (Norway) environment environment'.. Bloweretal Analysis of vehicle factors in serious truck crashes Trucks crashes from Factors previously found to be critical Crashes Driver violationsfout of service•. brakes out (2010) 2001 to 2003 reasons for serious crashes of adjustment or defective• (n=1,001 trucks & 963 crashes) (US) Cantor et al. Analysis of driver characteristics & crash history of truck Truck drivers with Numbers of prior crashes, drivers Crash rates Numbers of prior crashes.. . drivers under (2010) drivers using the Driver Information Resource data set crash data for a 2· under 25 with greater numbers of 25 with greater numbers of traffic held by the US Federal Motor Canier Safety year period ending traffic violations, vehicle maintenance violations... vehicle maintenance Administration September. 2007 violations, drivers' body mass index. violations.. . drivers' body mass index.. (n=560,695 drivers) us (Chen and Cross-sectional survey using a 7- point Likert scale to 239 Taiwanese pilots Perception of Safety Management Safety motivation (MO) SMS. ML and SE influence Mo• Chen, 2014) measure the effects of perceived SMS, morality from 5 Taiwanese Systems (SMS) Safety compliance (SC) SMS influences MO. SC & sp• leadership and self-efficacy and whether these factors air1 ines Morality leadership (ML) Safety participation (SP) MO influences SC & SP and moderates the are moderated by safety motivation. Pilots' self-efficacy (SE) effects of SMS, ML & SE on SC & sp• Safety motivation (MO) SMS and SE directly influence SC & sp• Cigularov et Cross-sectional survey to determine multi-level effects of Construction workers Safety communication, error Safety behaviours, injury and Safety communication•. error al. (2010) contractor error management climate and worker safety employed by 15 management (e.g., use PPP. follow pain management• communication on safety behavior, injury, and pain. contractors (n=235) safety procedures, report incidents. (US) assist others to work safely, take action to sto violations (Cui et al., Cross-sectional survey to test a model of safety 290 front line workers Hypotheses: Employee safety behaviour Hazardous environment - management 2013) management that links a hazardous environment. safety in a state-owned 1 Hazardous environment (HE) - Employee safety involvement commitment•.. climate and individual safety behaviours. using structural Chinese coal mining employee belief towards safety Management commitment - employee equation modelling techniques. company 2 HE - management commitment (MC) beliefs•.. 3 MC - employee beliefs Employee beliefs - safety behaviour".. 4a Individual beliefs - safety behaviour Safety involvement - employee beliefs.. • 4b Beliefs - safety involvement 5a MC - safety behaviour 5b MC - safety involvement (Dahl et al .. Cross-sectional survey using binary logistic regression 1108 offshore service Age Self-reported safety-compliant Age.. • 2013) analyses to examine factors that affect compliance to vessel workers Job experience behaviour Job experience.. safety procedures. working on 85 Safety climate Safety climate.. vessels chartered by Procedural vagueness Procedural vagueness..

66 Study Study design Participants Measured Outcome measures Significant predictors associated with safety outcomes one Norwegian oil company Darby et al. Online risk assessment and cross-sectional survey about Employees of a large Driver attitudes, behaviours, Crashes Aggressive personality*... age•••. having (2009) driver demographics, driving history and driving telecommunications knowledge. hazard perceptions, age. licence check*** behaviour and knowledge against employee crash company (n=16,004) personality trait records. UK (DeJoy et al., Cross-sectional survey to test a model of organisational 1.723 employees of a Core values OSH policies & procedures Core values predict OSHP&P• 2010) social exchange influences on safety using a structural large US national OSH policies & procedures (OSHP&P) (OSHP&P) OSHP&P predict commitment, climate & equation modelling approach and Chi-square to test the retailer Perceived organisational support Perceived organisational support Pos· model fit. (POS) (POS) POS mediates effects of OSHP&P on Safety climate Safety climate commitment & climate• Organisational commitment Organisational commitment Commitment predicts vitality, withdrawal & Worker vitality perception of safety• Withdrawal behaviour Climate predicts perception of safety and Perceived safety at work accidents• (Self-report) accidents Feyerand Cross-sectional survey of drivers about work, rest and Truck drivers (n=989) Work scheduling Perceptions and experience of Work scheduling** Williamson fatigue (Australia) driver fatigue 1995 (Friswell and Cross-sectional survey of drivers about safety concerns 321 Australian light Road and access factors Self-reported injuries Organisation of work*** Williamson. and comparisons with self-reported injury causation transport vehicle Organisation of work Vehicle-related hazards**• 2010) factors. Subgroups of injured and non-injured drivers' drivers Vehicle-related hazards Interpersonal conflict• responses were compared using Chi-square and Mann Interpersonal conflict Whitney tests. Interviews and group discussions were also used. Binary logistic regressions were also rformed. Fogarty and Cross-sectional survey on safety climate and effects on Aircraft maintenance Perceptions of management attitudes. Self-reported violations of safety Management attitude**. worker attitude••. Shaw (2010) safety behaviour personnel (n=308) own attitudes. group norms and procedures and intentions to group norms.. . work pressure** (Australia) workplace pressures as they affect violate safe behaviour. Golob and Cross-sectional survey to model the links between work Truck drivers (n=402) Work demands, pill-taking, self Pill-taking, self-imposed schedules•. Henshe r demands and pill-taking and speeding (Australia) imposed schedules, carrying self-imposed schedules. carrying perishable goods*. (1994) perishable goods, company size speeding, speeding• traffic fines (Guestet al., NSW truck crash data was analysed using distance 12,501 Australian Age of professional driver Crashes of heavy articulated Compared with drivers aged 45-54. age 2014) travelled as exposure and examining age of driver as a male heavy truck trucks groups 18-20, 21-25 and 26-34 have higher risk factor. Negative binomial regression modelling was drivers Crashes of heavy rigid trucks rate ratios for crash involvement... but used to estimate annual crash incidence rate ratios for (1 Jan, 1999-31 Dec. 2006) there is no difference for age groups 55-64 male drivers in various age groupings by vehicle type. nor 65+ (Huang et al. . Cross-sectional survey to determine important perceived Management commitment to safety, Safety climate and self-reported Management commitment to safety**, 2006a) mitigating factors to self-reported injury risk in workplaces return-to-work policies, post-injury injury incidents return-to-work policies.. . post-injury administration, employee control and administration••. employee control** and safety training safety training••

67 Study Study design Participants Measured Outcome measures Significant predict01'8 associated with sat outcomes (Huang et Cross-sectional survey to design and test a safety climate 2,421 mobile and Organisational safety climate (OSC) Self-reported: OSC predicts near misses and incidents** al., 2013b) scale through exploratory factor analysis and confinnatory remote workers from Safety pro-activity Safety behaviours GSC predicts near misses, incidents, and factor analysis methods two large US electric General training Injury outcomes: missed days• utility companies Trucks & equipment 0 Near misses OSC predicts safety behaviour •• Field orientation 0 Recordable incidents GSC predicts safety behaviour •• Financial investment 0 Vehicle accidents Schedule flexibility Objective data Group safety climate (GSC): - Hard braking Supervisory care - Missed work days due to - Participation encouragement injury - Safe strai hi talk Kath et al. Cross-sectional survey to rate a scale of safety climate Employees of a Trust in organisational Safety climate, job satisfaction, Trust in organisational communications**•, (2010) indicators grocery store chain communications, perceived safety motivations, turnover perceived management attitudes**• (n=599) (US) management attitudes intentions (Kemp et al., Cross-sectional survey to examine stressors on truck 435 US truck drivers lime pressures Physical fatigue lime pressure to stress** 2013) drivers and impacts on road safety Stress Emotional exhaustion Stress to physical fatigue** Altitudes to regulation HOS violations Stress to emotional exhaustion** Car driver behaviour Safety compliance Physical fatigue to attitude to safety compliance** Emotional exhaustion to attitude to safety compliance** Attitude to safety compliance - attitude to regulation** Attitude to safety compliance - exceed HOS** (Lu and Tsai, Cross-sectional survey testing 22 safety climate items as 608 seafarers Safety policy relationship with safety Safety climate model Safety policy positively relates to safety 2010) attributes of 3 factors (safety management, perceived working on 124 management; Self-reported behaviour management• supervisory behaviour and safety policy) based on vessels from 13 Safety policy relationship with Safety policy positively relates to supervisory assumptions and values relied upon in different studies. countries perceived supervisory behaviour safety behaviour• Exploratory factor analysis found that the 3 factors Safety policy relationship with Safety policy positively relates to seafarers' accounted for 62% of the total variance. However, the seafarers' behaviour; Safety behaviour* confirmatory factor analysis resulted in 2 items being management relationship with Safety management positively relates to dropped from safety policy and 3 items from safety supervisory behaviour; Safety supervisory behaviour• management A structural equation model was then used management relationship with Perceived supervisory safety behaviour to test the hypothesized safety climate model. seafarers' behaviour; and Perceived positively relates to seafarers' behaviour• supervisor behaviour relationship with seafarers' behaviour. (Lu and Cross-sectional survey of perceptions of safety leadership 336 workers at 5 Safety leadership with respect to: Safety compliance Safety leadership with respect to: Yang, 2010) and self-reported safety compliance and safety container terminals in safety motivation Safety participation safety motivation - compliance** participation. A confirmatory factor analysis was Taiwan safety policy safety motivation - participation** undertaken as well as !-tests (for convergent validation) safety concern safety policy-participation• and structural equation modelling (for discriminant validity safety concern -compliance** assessment). safety concern - participation**

68 Study Study design Participants Measures Outcome measures Significant predict01'8 associated with outcomes Mayhew and Cross-sectional structured interview study with Long-haul truck Company size, economic pressure, Wori

69 Study Study design Participants Measures Outcome measures Significant predictors a88ociated with aat outcomes Williamson Analysis of 2 cross-sectional surveys of Australian long Australian truck Pay system Driver fatigue. use of stimulant Payment by results versus hourly (2007) distance truck drivers about predictors of fatigue and drugs payment ••• stimulant drug use.

(Williamson and Cross-sectional self-report survey to assess whether Driver payment (trip or time) Self-reported fatigue Usual experience: Friswell. 2013) payment systems for driving and non-driving activities Waiting/paid for waiting Longer working hours Trip based payment• can predict driver fatigue risk. Data were collected on Sleep prior to trip Greater distance driven Not paid for waiting/queuing• driver characteristics, fatigue risk experience and driver Driver experience and age Last round trip experience: payment method. Not paid for waiting/queuing• Amount of sleep prior• HV driver experience• Willsetal. Cross-sectional survey to cross-validate the factor Employees from 3 Communication and procedures. work Self-reported driver error, pre Communication and procedures.. . work (2005) structure of an existing survey measure of safety climate, Australian pressure. management commitment. trip maintenance behaviours, pressure.. • . management commitment•. in different workplace settings. working relationships, driver training & driver distraction & violations driver training... & safety rules... safety rules

Zohar (2002) Cross-sectional survey to analyse effects of leadership Safety climate. leadership styles. Objective data recorded Leadership style.. . assignment of priOOty style, concern for safety and assigned priority to safety in a metal assigned safety priorities injuries of safety• (from higher management) processing plant (n=411) (Israel) (Zohar et al., Cross-sectional survey and prospective in-vehicle heavy Long distance truck Safety climate, dispatcher leadership. Safety climate, driving safety, Leadership to safety climate.. 2014) braking measurement to examine safety climate drivers (n=3.578) of work ownership. driving safety Objective data: heavy-braking Ownership to safety climate.. measures against safety behaviours one large US (safety incident) Safety climate to driving safety.. company Driving safety to (-) hard braking* Reported p-values: •p D< 0.05. ..P D< 0.02 ...P D< 0.002

3.3.3 Studies of the effects of organisational interventions on safety outcomes The review found nine (9) studies in which organisational interventions were evaluated for effects on safety. Table 3.3 presents findings of studies that focused on specific workplace interventions. Most (8) were longitudinal intervention studies. Six (6) studies (67%) were in the fleet/driving context. Four (4) of the nine studies (44%) involved heavy vehicle fleets. Four (4) out of the nine (44%) used crash or crash costs as the outcome. Five (5) studies (56%) examined changes in behaviour, attitude or climate following an intervention.

Review of the findings of the studies suggests that safety training and education is likely to be effective in improving safety outcomes as this was found in three (3, 33%) of the five studies. Four (4, 44%) studies found that worker participation in WHS was associated with improved safety outcomes. Other characteristics associated with improved safety in one study include incentives, pay systems and pay rates, rest time and vehicle safety technologies.

71 Table 3.3 Studies of organisational interventions and their effects on safety outcomes Study Study design Participants Measures Outcome measures Significant predictors associated with good safety outcomes Gander et Longitudinal pre and post fatigue management Heavy vehicle Driver fatigue training Self-reported fatigue Fatigue safety strategy training••• al. (2005) training quiz and follow-up survey to evaluate drivers (n=72) management strategies (NB p.value was calculated as it wasn't included in the paper.) whether training is an effective fatigue risk (New Zealand) at home and work management measure. Gregersen Longitudinal intervention study testing the impacts of 5 groups (including Driver education, safety Crash costs Driver training•. safety promotion• incentives•. group et al. four types of intervention to reduce light vehide work 1 control) of 900 promotion, incentives, group discussion• (1 996) related crashes. The interventions included: driver drivers in a discussions education. safety promotion. incentives and a group Swedish discussion on risk management telecommunications organisation (n= 4500 Sweden (Kines et Intervention study involving four problem-solving 14 small (10-19 Application of a participatory (Process) Safety Integrated safety management approach (affective al.. 201 3) dialogue meetings with ownersfmanagers to gain employees) Danish behaviour-based injury participation, trust. commitment improvement, safety knowledge and safety commitment. reflect on leadership role, prioritise metal industry prevention approach affective commitment involvement) .. safety, increase safety communication. identify safety companies: 6 integrated with safety culture (Effect) Safety problems, establish remediation activities and intervention and 8 initiatives knowledge, priorities. and initiate, follow up and evaluate tangible control companies involvement, attitudes, safety activities taking an integrated approach to leadership. and safety safety management (combining behaviour-based and representative culture change. Pre- and post surveys and comparisons between intervention and control com anies. (Nielsen et An intervention study to test the effectiveness of a 105 workers at 2 Workshops (3) involving Safety leadership Intervention - al.. 201 3) participatory safety problem-solving workshop small (2049 supervisors and workers on Safety knowledge Safety leadership, knowledge & safety rep commitment.. (for intervention in terms of safety culture improvement employees) Danish safety performance, Safety involvement the metal company only) metal and wood discussions on safety issues. Safety behaviour Safety involvement & behaviour" (metal company only processing and safety management and Safety rep commitment companies, and 2 leadership control companies Rodriguez Longitudinal study to determine whether a pay rate Employees of one Pay increases Crashes, driver turnover Pay increases• et al. increase of 39.1% would result in crash reductions. trucking firm (2006) (n=2.368) (US) Salminen Longitudinal experimental study involving a Employees of 2 Driver discussion group Crashes The driver discussion group intervention•.. . anticipatory driver (2008) comparative test of a group discussion method (of electricity method, driver training training• initial reduction but increase over a 3-year period reducing injury risk) and anticipatory driver training to companies (n= reduce risk. 351) (Finland)

(Wang and Experimental study involving measurement of driver 33 Chinese coach Driving hours and resting Speed perception Impact on driving performance Pei, 2014) sleepiness and performance in on-road driving by 3 drivers from one times Depth perception - after 2 hours: attention allocation value decreased* and groups of commercial drivers: one measured after 2 company Attention allocation choice reaction times* and fatigue increased* hours, one after 3 hours and one after 4 hours. The value - after 3 hours: attention value*, reaction time** and fatigue study used instruments that tested drivers’ speed Choice reaction time increased*** and incorrect action judgements increased* perception, depth perception, attention allocation, Number of correct light - after 4 hours: speed perception*, attention**, reaction reaction time and action-judgement. The Stanford reactions time***, reaction judgements*, fatigue*** all showed Sleepiness Scale (SSS) was also used as a Number of incorrect changes subjective measure of sleepiness. action judgement Impact of rest time on recovery SSS scores - after resting for 15 minutes after 2-drives all performance measures were as before driving the 3 to 4-hour drives and 4 or more hour drives required a 30-minute rest to fully recover Wouters Longitudinal field trial used a ‘matched 840 vehicles in 11 In-vehicle data recorders and Crashes The installation of in-vehicle data recorders and driver and Bos experimental/control group” design with intervention Dutch commercial driver feedback devices. feedback devices*, but the effects varied considerably (2000) to assess whether in-vehicle feedback devices would fleets (270, plus between types of fleets (taxis vs trucks) and management reduce the number (and/or severity) of road traffic 570 control feedback practices were not measured. crashes. vehicles) (Zohar and Randomised field study to test a 12-week supervisor 342 workers in a Supervisor feedback Higher safety climate Worker-supervisor feedback intervention (all outcome Polachek, to employee communication intervention and worker mid-sized Israeli intervention Higher safety behavior measures) ** 2014) feedback sessions aimed at improving safety climate, heavy level using pre- and post-survey data and independent manufacturing Lower perceived safety audit data. company, divided workload into 26 work teams Higher level of with one supervisor perceived teamwork divided into Higher safety audit experimental and scores control groups Reported p-values: *p < 0.05 **p < 0.02 ***p < 0.002

73 3.4 Summary of results

Combining the results of the 68 studies included in this review shows some consistent findings across studies (see Table 3.4). Across the three types of studies described previously, it is clear that the evidence for a link with safety outcomes is most consistent for management commitment/safety climate, supported in 30 studies (46%) across all three groups of studies, with five of these studies (17%) using objective safety outcome data. Confirmatory evidence was found for worker participation in WHS that was shown in 21 studies (32%). Vehicle and work conditions were found to have links with safety outcomes in 13 studies (19%). Safety training was also commonly supported by evidence from 12 studies (18%) across all three types of studies. Work scheduling and journey planning was supported in 11 studies (16%) including all three types of studies. Financial performance, pay systems/rates and/or unionisation were confirmed in 8 studies (12%) combining the three different types of studies. Incentives for safety were researched in 7 studies (10%), including organisational and intervention studies.

The link with safety outcomes was found in the organisational and individual studies but not for intervention studies, for safety management accreditation, safety policies/procedures/enforcement, risk analysis and corrective actions, size of organisation, worker characteristics, hiring and retention, and prior safety incidents. Each characteristic listed in the summary, Table 3.4 were confirmed across at least two studies, and one characteristic, management commitment/safety climate, was found a significant safety management characteristic in 30 studies.

74 Table 3.4 Summary of the number of studies showing significant relationships between the characteristic and safety outcomes for organisation level, individual level and intervention studies

Characteristic Organisation- Individual- Intervention Total level studies level studies studies (n=68) (n=25) (n=34) (n=9) Management commitment/safety 8 21 30 cl imate Worker participation in WHS 8 8 5 21 /communication Vehicle conditions/ physical or 4 8 13 psychosocial work environment Safety training 7 2 3 12 Scheduling/journey 4 6 11 planning/pressure Safety 3 5 8 policies/procedures/enforcement Financial performance/pay systems 3 4 8 rates/unionisation Risk analysis/corrective actions 7 8

Incentives 6 7 Size of organisation/number trucks/ 3 3 6 Freight type Safety management systems 4 6

Worker characteristics: driver 4 4 attitudes/behaviour/age/health Hiring/retention/return to work policy 3 4

Accreditation 3 3

Prior safety violations, 2 2 crashes/incidents

In all, 38 out of the 68 studies (56%) were undertaken in the heavy or light vehicle transport context, and of these 28 studies (41 %) were undertaken in the heavy vehicle transport sector. Some of these studies examined characteristics that are only relevant in fleet safety, such as vehicle conditions and vehicle safety technologies. Characteristics that were studied in other industry sectors were also examined in studies into heavy or light work-related driving safety.

Many of the heavy vehicle transport safety studies looked at more constrained sets of characteristics rather than safety management systems or culture. The most prevalent predictor studied in the road freight transport context was financial wellbeing or remuneration systems or rates with all 8 studies (29% of the 28 studies). This may be due to increasing attention, particularly by Austra lian researchers on systemic risk factors in the road freight transport sector.

75 Other characteristics that showed the most consistent evidence for links with safety outcomes in the transport sector were size of fleet or type of freight with these variables tested in 6 studies (21%) in this sector. Scheduling/journey planning/ work pressures as well as commitment/safety climate were the focus of five (18%) each, and four (14%) examined effects of vehicle conditions/work environment. Three (3) studies (11%) examined worker characteristics.

While management commitment, leadership and safety climate were examined in the most studies (n=30) in the review of the total 68 studies in total (44%), these were examined in only five (5) studies (18%) in the heavy vehicle transport sector. The review found 3 studies (11%) each on worker characteristics and worker participation, whereas there were seventeen (17) studies (25%) out of the total 68 studies examined that looked at worker participation. Accreditations/SMS, hiring/retention, safety training and prior violations/crashes were also the subject of 2 studies each. Moreover there were twelve (12) studies (18%) on safety training overall but only two (2) studies (7%) in the heavy vehicle transport sector. The review found among the heavy vehicle transport studies only one (1) study each focused on safety policies, risk analysis or incentives.

Table 3.5 provides a summary of the evidence based safety management characteristics found in studies involving the heavy vehicle transport industry in this review.

76 Table 3.5 Summary of the number of heavy vehicle transport safety studies showing significant relationships between the characteristic and safety outcomes for organisation level, individual level and intervention studies

Characteristic Organisation Individual Intervention Total level studies level studies studies (n=28) (n=11) (n=13) (n=4) Financial performance/pay systems 2 5 8 rates/unionisation Size of organisation/number trucks/ 3 3 6 Freight type Scheduling/journey 3 5 planning/pressu re Management commitment/safety 3 2 5 cl imate Vehicle conditions/ physical or 2 4 psychosocial work environment Worker characteristics: driver 3 3 attitudes/behaviour/age/health Worker participation in WHS 2 3 /communication Safety or quality management 2 2 accreditation/SMS Hiring/retention 2

Safety training 2

Prior safety violations, 2 2 crashes/incidents Freight type

Safety policies/procedures/enforcement Risk analysis/corrective actions

Incentives

3.5 Discussion of literature findings

The objective of this research was to inform the development of an evidence-based SMS for trucking that results in reduced crash outcomes. The current chapter attempted to identify some of the characteristics of SMS that are most likely to achieve improvements in safety in heavy vehicle transport through a review of the existing empirical research literature.

Currently, one of the problems with the SMS approach is that any number of plausible links between safety and work or management practices might be proposed for inclusion within the coverage of an SMS, but there is little evidence-based empirical guidance about which factors are most important to manage safety. This finding is consistent with previous research, namely a very broad pri or review of 4,837 studies by Robson et al. (2007) that found very little definitive peer-reviewed evidence for th e

77 effectiveness of workplace safety management systems in general. A more focussed synthesis of the results of the small number of studies reviewed in this thesis suggested that it is likely that some safety management characteristics may be able to produce good safety outcomes in terms of incidents and injuries.

Moreover, approaches to safety and injury management have not been shared between the WHS and the road freight transport sector. Research into safety management of work related driving in light vehicles (less than 4.5 tonnes) has found that there is also widespread neglect of this area of injury prevention (Murray, 2009; Newnam and Watson, 2011).

Despite this, auditable forms of safety management systems, sometimes termed accreditation or alternative compliance programs, have been developed and implemented. A range of these programs currently exist, such as a fire safety management system (Santos-Reyes and Beard, 2002), WHS certification (Zwetsloot et al., 2011), safety quality assessment systems (Verlinden, 2002), and OHSAS18001 19 and AS4801 20 . However, although there is some evidence that certification to OHSAS18001 (Abad et al., 2013) and to NHVAS (Baas and Taramoeroa, 2008) may improve safety, there are conflicting research results (Jansen and Dikranian, 2009). Most evaluation studies of this type suffer problems of self-selection to the accreditation program, i.e. only good performing companies join. The elements of the Australian industry schemes, notably TruckSafe (Australian Trucking Association, 1991), and the National Logistics Safety Code (Australian Logistics Council) appear to have a safety management focus, but neither of these have been evaluated. In short, while they have shown promise, there is no clear evidence of their ability to produce safer outcomes. i.e. reduced crash outcomes.

This strategic literature review found limited research to date that provides the beginnings of an evidence base for a set of safety management characteristics that have been associated with improved safety outcomes. In fact, only 68 original research studies were found that provided evidence of statistically significant links between safety management characteristics and safety outcomes. Of these, only 25 studies (37%) focussed on comparing differences at an organisation or company level. Moreover, many of the studies found in the review used self-reported safety outcomes rather than objectively obtained crash or injury statistics, raising questions of the validity of safety outcome measures.

The majority of studies into the characteristics that predict improved safety outcomes used the individual person, group or vehicle as the unit of study (35). While this evidence is useful, it may not be possible to understand whether the relationship between characteristics and safety outcomes operate at the organisation or system level or whether they are simply due to individual differences. For example, the findings that employee age (Cantor et al., 2010; Darby et al., 2009), personality

19 See https://www.saiglobal.com/Assurance/ohs/OHSAS18001.htm 20 See http://www.healthsafety.com.au/as4801/

(Darby et al., 2009) or body mass index (Cantor et al., 2010) are related to individual safety outcomes do not explain systemic differences between company safety management practices. On the other hand, characteristics like safety and return to work policies that were also related to individual safety outcomes in six studies could be due either to the influence of organisational policies that would influence most employees, or to the response of particular individuals.

Despite these limitations the review highlighted some characteristics that have shown evidence from a large number of studies supporting their preliminary inclusion in a safety management system (SMS) suitable for heavy vehicle transport operations. In particular, management commitment/safety climate, worker participation, vehicle conditions/physical or psychosocial work environments, safety training, and scheduling and journey planning were confirmed as statistically significant predictors of safety outcomes in ten or more studies each including heavy trucking as well as other industry sectors. In addition, there was support for links with safety outcomes in at least eight studies each for safety policies/procedures/enforcement, pay systems/rates, and risk analysis/corrective actions. These results provide direction for what are likely to be effective management targets within an SMS. Interestingly, these characteristics cover both cultural aspects of organisations like management commitment and worker participation and policy and practice-related aspects like safety training and work scheduling.

Importantly, most models of safety management systems have argued for the importance of management commitment and worker participation as components of SMSs (Guldenmund, 2000; Hale et al., 1997; Hale and Hovden, 1998) although the authors often admit that the evidence is lacking. This review of relevant research adds some much needed confirmatory evidence to theories of effective safety management. However, caution should be used as there are still many gaps in the research and this review was limited to the safety management items that happened to be of interest to the author of this thesis and do not confirm whether or not these items are more important than others that may be studied in future. The analysis in this review only looked at evidence of characteristics shown to be important across studies, it did not look at characteristics that have been tested and shown not to be important nor does it comment on characteristics that have not been tested at all. The value of this review is that it distils those characteristics that have been tested on an evidentiary basis and demonstrated to be related to safety outcomes.

Second, many of the studies reviewed involved pragmatic, but not strong research designs. For example, out of the 65 studies, 45 (69%) were cross-sectional surveys. These can show differences between companies in terms of relationships between characteristics and safety outcomes, but it is difficult to infer a causal relationship without longitudinal data or effective control groups. Nonetheless, some of the cross-sectional research findings were quite robust. For example, two cross-sectional industry surveys conducted 10 years apart, found similar links between remuneration systems and safety outcomes (Williamson, 2007). Conversely, there can be practical problems involved in

79 ensuring that stronger longitudinal designs remain valid when the workforce is highly mobile or untenured.

Third, only 24 (37%) studies used objective crash or incident outcomes as the safety performance measure. The rest used self-reported behaviours, perceptions and incidents. Self-report measures are uniquely able to measure many variables of interest, such as subjective experiences of fatigue or stress, and symptoms of injury and ill health. They also permit access to, and can affect, people’s motivations for engaging in safe and unsafe behaviours. However, self-reports can be affected by intentional or unintentional biases and can suffer common methods bias when the characteristic of interest is also self-reported. Apparently ‘objective’ measures can also be subject to sources of inaccuracy. For example, insurance claims data can be affected by reporting practices or company claims policies (Quinlan and Bohle, 2004). In those circumstances where multiple measures are available, the replication of findings across objective and subjective measures provides important converging evidence for effects. In the current review, finding similar results with subjective and objective safety outcomes, such as the studies linking characteristics like management commitment and safety training with safety outcomes, further strengthens the evidence that these characteristics should be considered for inclusion as components of effective SMS. Characteristics that showed converging evidence from objective outcome measures and self-report surveys included worker participation or opportunity for safety input (Geldart et al., 2010); (Arboleda et al., 2003) and company size (Monaco and Williams, 2000; Moses and Savage, 1994). Two characteristics appeared in studies with objective or self-report outcomes and in evaluation studies. These were safety training (Arboleda et al., 2003); (Gregersen et al., 1996; Wills et al., 2005) and pay-related issues (Belzer et al., 2002); (Williamson, 2007); (Rodriguez et al., 2006).

In this review, the strongest research designs were longitudinal cohort studies of organisational interventions, although some of these studies were weakened, as they did not include control groups. There were only nine cohort/longitudinal studies: four associated with heavy transport and two with light vehicle fleets. Three of these studies evaluated safety-training interventions and, like seven of the organisation level studies, found safety improvements including, for two of the studies, on actual, rather than self-reported crash outcomes. Four studies evaluated worker participation in WHS in the form of driver discussion groups and, and in all cases found benefits for crashes and crash costs, again supporting the findings of six organisation level studies. Even though the intervention studies had a much narrower focus on characteristics and differed markedly in study design compared to the organisational and individual level studies, the nature of the findings was consistent between the three types of studies. In combination, the three types of studies were able to identify characteristics that warrant further research consideration.

In the context of the increasing demand to improve the standard of literature reviews, which is at the heart of evidence-based public health, this review could be seen as lacking. Literature reviews are an

essential step in research, policy and practice as, at least in theory, they help to tell us where we are with respect to an issue or problem, what we might do about it and what else needs to be done. There is little doubt that the standards for literature reviewing have risen considerably with the introduction of more rigorous guidelines for synthesising research evidence (Higgins and Green, 2008; Mulrow, 1994). Unfortunately, however, in many areas, particularly of public health, the benchmark for systematic reviewing is high and potentially unachievable (Dixon-Woods et al., 2005; Mullen and Ramirez, 2006). This is certainly true for many workplace and road safety issues. At this stage in research on what makes an effective SMS, we are faced with an exceptionally broad set of possible characteristics and a large number of studies, but mostly comparatively weak designs. This review was an attempt to answer the question of the most likely characteristics or practices to produce demonstrable differences in crash and injury rates using the key principles of systematic reviewing. These include the need to be comprehensive, systematic and transparent, where being systematic in particular involves clear inclusion criteria, systematic searches and selection of studies based on predefined criteria. The diversity of studies and weak study designs mean, however that it was not reasonable to attempt to comply with the full recommendations for systematic reviews and conduct a statistical meta-analysis on the studies generated. Rather, for this review it was necessary to synthesise the results through a narrative review accompanied by a vote counting approach (Verbeek et al., 2012). The potential disadvantage of this method is that findings of no effect may be spurious due to low statistical power. To avoid this problem, as acknowledged above, the interpretation of the results of this review only included studies where statistically significant results were found. With regard to research methods, this review has drawn attention to the need for better- controlled designs and greater focus on the implementation and evaluation of safety management practices.

As highlighted earlier, the research on SMS is quite limited both in amount and in scope and that applicable to heavy vehicle transport is even more limited. Further research is needed to help to expand the scope of studies on predictors of safety outcomes in work settings like heavy trucking. This review has revealed some characteristics that have reasonable evidence for a place in an effective safety management system, including safety training, worker participation and restraint on the use of payment systems to emphasise productivity. For other characteristics questions remain despite some existing research. Questions remain for example on the role of management commitment, found to be related to safety in many studies, but all involved self-reported outcomes. None of the studies with objective outcomes included management commitment factors. Clearly further research is needed to confirm the importance of this characteristic.

Other characteristics that have attracted some research but for which questions remain include the role of risk analysis, hiring practices, safety accreditation systems and the influence of company financial performance.

81 Almost certainly many other factors that might influence safety outcomes have simply not been studied. For example, there has been no little or no research on the impact of regulation on safety management in heavy vehicle transport. The transport and logistics chain from customers to delivery drivers can involve a complex web of contracting arrangements which can produce economic incentives that are antagonistic to safety, but that are difficult to control (Rawling and Kaine, 2012). Australian regulatory authorities have responded by introducing chain of responsibility principles to transport regulation, and more recently, an independent process of setting driver payment rates, but the effects on safety management at the level of individual transport companies is not known.

Overall, the literature review identified a number of characteristics that alone or in combination with other characteristics may be able to predict crash and/or injury outcomes. The ultimate measure of safety is the frequency and severity of injury or incidents and their costs. However, the relationship between organisational or management characteristics and safety outcomes is often indirect. That is, the establishment of an environment conducive to safe behaviour – whether these behaviours are proximal or distal from the risk event – does not guarantee safe behaviour will occur (e.g. wearing a seat belt, choosing a safe vehicle, operating a vehicle at safe travel speeds, etc.). Yet very few studies (Fernandez-Muniz et al., 2007a; Golob and Hensher, 1994) have sought to examine the complexities of the relationship between safety management practices and safety outcomes. Clearly much more research is needed to understand how characteristics interact to promote safety in workplace settings. Unfortunately, this is a step-by-step process in identifying the important contributing factors from the confounding ones. This review has attempted to use a systematic approach to begin this process.

3.6 Conclusions of the strategic literature analysis (Study 1)

This study set out to distil from the scientific safety literature safety management characteristics that have been shown to have links to safety outcomes. The outcome measures found in previous safety research include, safety incidents, including crashes, injuries, self-reported near misses, safety errors, violations/compliance, safety and risk perceptions, safety participation, and safe/unsafe behaviours.

It revealed safety management practices and organisational characteristics that have been associated with lower levels of incident and injury risk. While there is not yet a single set of characteristics that have been reliably shown to underpin effective safety management, there are indications that some safety management and organisational features are worthy of further research in heavy vehicle transport operations, particularly given the high crash rates and injury risk associated with heavy vehicle drivers’ work as discussed in Chapter 1.

The 68 studies examined in this strategic review found a range of organisational features that are associated with good safety outcomes. Some of the characteristics that were tested in these studies, were tested in as many as 30 studies in various types of companies. The prevalence of items studied

in safety research with statistically significant findings lends strength to the validity of these characteristics. The specific safety management characteristics that have demonstrated links with safety behaviours and safety outcomes in a number of workplace settings in two or more studies, in order of highest to lowest relevant number of studies, that found a significant link with safety outcomes are:

Management commitment/safety climate, Worker participation in WHS /communication, Vehicle conditions/ physical or psychosocial work environment, Safety training, Scheduling/journey planning/pressure, Safety or quality management accreditation/SMS, Safety policies/procedures/enforcement, Financial performance/pay systems rates/unionisation, Risk analysis/corrective actions, Incentives, Size of organisation/number trucks/ freight type, Worker characteristics: driver attitudes/behaviour/age/health, Hiring/retention/return to work policy, and Prior safety violations, crashes/incidents.

Further, examining the findings of the 28 studies that were undertaken in the heavy vehicle industry, the characteristics that showed evidence of links to safety outcomes in 2 or more studies in order of most to least relevant number of studies with significant results are:

Financial performance/pay systems rates/unionisation, Size of organisation/number trucks/ freight type, Scheduling/journey planning/pressure, Management commitment/safety climate, and Vehicle conditions/ physical or psychosocial work environment Worker characteristics: driver attitudes/behaviour/age/health Worker participation in WHS /communication Safety or quality management accreditation/SMS Hiring/retention Safety training Prior safety violations, crashes/incidents

Finally, the findings from this review (Study 1) provided a sound basis for selecting items to test in a survey of heavy vehicle road freight transport operators (Study 2). Chapter 4 will describe the

83 development of that survey, the survey methods including the administration of the survey to managers from groups of lower- and higher-claiming companies and will describe the survey findings.

Chapter 4: Survey of Characteristics Distinguishing Between Better and Poorer Safety Performers (Study 2)

This chapter discusses the findings of the second study, a survey of managers in lower truck insurance claiming companies and managers in higher truck insurance claiming companies. This survey aimed to identify the management characteristics that distinguish between lower- and higher-claiming companies.

4.1 Introduction

As discussed in Chapter 1, the trucking industry and specifically heavy vehicle driving is a high safety risk enterprise. This industry sector with its low profitability levels, intense competition, concentration of buying and selling power, and complex web of contracting and subcontracting brings particular challenges to for managing safety.

Unfortunately, current approaches to reducing the problem of road trauma in trucking, including those developed from a road safety perspective, have not been very successful. New strategies are needed to make an impact on the problem. In Chapter 2, the efficacy and applicability of safety management approaches in other sectors, particularly work health and safety (WHS) were examined. Ideas around Safety Management Systems that take a broad view of the work setting and the sources of solutions to safety issues show promise for the trucking industry, but they have not been implemented in any significant way.

Although a range of risk factors have been identified for HV driver injury (Department of Transportation U.S., 2006; Loeb and Clarke, 2007; Lueck, 2011; Parker et al., 1995; Williamson, 2005, 2007), organisational practices that may be used to manage the risks to drivers have received surprisingly little research attention. The first study undertaken in this thesis (Chapter 3), was a strategic review of the empirical research that provides scientific evidence of links between management characteristics and safety outcomes. A total of 28 studies, particularly focused in the road freight transport industry found some links between specific characteristics and safety outcomes. Nevertheless, across all industry sectors there has been little consolidation of the evidence to understand which characteristics distinguish companies that have good safety records from those with poor safety records in general, and certainly no evidence on what factors might be important in the trucking industry.

Overall, the review of the research literature, covered in Chapter 3, revealed that there are some important organisational characteristics to examine when attempting to find the most important mix of safety management characteristics in heavy vehicle transport companies. These characteristics may be grouped under twelve topic headings:

1. Size of organisation and freight task; 2. Profitability;

85 3. Vehicle selection, equipment and maintenance; 4. Journey risk assessment and work scheduling; 5. Driver recruitment and tenure; 6. Pay and conditions; 7. Policies and safety accreditation; 8. Incentives for safety performance; 9. Training and information; 10. Driver participation in workplace health & safety; 11. Incident analysis and record keeping; and 12. Driver management and discipline.

Using these topics a survey of Australian trucking companies was designed, administered and analysed. The aim of the study was to identify distinguishing self-reported management and organisational characteristics of better safety performing heavy vehicle operating companies and poorer safety performers as indexed by truck insurance claims over a three-year period. This study is presented in this Chapter.

4.2 Methods for Study 2

4.2.1 Design The overall design of this study involved a comparison of the reported practices in good and poorer safety performing heavy trucking companies. A sample of trucking companies operating in Australia were categorised as good and poorer performing on the basis of the number of safety-related insurance claims made per truck. Managers of these companies were then surveyed in 2011-2012 to identify whether any of a range of safety management characteristics identified as potential influencers of safe performance in previous research discriminated between companies with good and poorer safety performance.

Safety performance was indexed using vehicle damage insurance claim rates. Finalised heavy vehicle claims data for the period 2007-2009 were available from a major insurer of heavy vehicle fleets in Australia. The claims period was selected to i) include the most recent years with a high percentage of finalised claims (mean per year = 98.4%), ii) cover a sufficiently long period that the claim rate would have some resilience to atypical peaks in claims, and iii) be as recent as possible to minimise the possibility of significant organisation changes since the claim period. Organisational claim rates per heavy truck for the period were calculated using claims where at least some fault was attributed to the insured (i.e., excluding claims for natural occurrences, 100% third party fault or unattributed fault). Companies were selected from this subset of claims to be invited to participate in this study on the basis of two criteria of interest: their claim rate and their heavy vehicle fleet size. Good performers were defined as those with the lowest 30% of claim rates (rate = 0 claims per truck) whereas poorer performers were defined as the highest 40% of claim rates (rate > 0.17 claims per truck).

Fleet size may influence the extent and sophistication of the safety management practices that an organisation adopts (Eakin et al., 2010; Knipling et al., 2011; Mitchison and Papadakis, 1999) and so it was originally intended to also examine the effect of fleet size. Small and large fleet size groups within each claim rate group were created by selecting only from the largest 40% of companies and the smallest 40% of companies. This meant that large companies were empirically defined as those with 14 or more heavy trucks and small companies were defined as 8 or less heavy trucks. Companies with only one or two trucks were excluded on the basis that they were so small they would be unlikely to adopt formal safety management practices.

Use of safety management practices was self-reported and included i) vehicle acquisition, maintenance and disposal policies and procedures, ii) driver recruitment, payment and training arrangements, iii) other safety policies, iv) participation in accreditation schemes, v) journey and schedule planning practices, vi) driver participation in Work Health and Safety management, vii) driver monitoring and incentives, and viii) safety data recording.

4.2.2 Participants Fifty organisations meeting the claim rate and size criteria participated in the study. They were drawn from across Australia and were recruited in two ways, either via their insurer or via their telephone listing. First, a major insurer extended an invitation via its insurance brokers to all clients that met the selection criteria to participate in the study. The brokers were asked to either forward the invitation or to agree (or not) to allow researchers to directly invite their client to take part. Unfortunately, this strategy resulted in very few companies volunteering (n=17). To increase the sample, additional companies were recruited from a random sample of transport companies listed in the Australian Yellow Pages. Only those companies meeting the same claim rate and company size criteria that were applied to the insurance sample were included in the final sample. In total, there were 20 low-claiming companies (9 small and 11 larger), and 30 higher-claiming companies (12 small and 18 larger) in the final sample.

4.2.3 Materials A questionnaire with some 60 questions was prepared, based on the twelve topics identified in the Study 1 strategic literature review. This questionnaire was tested by consulting five industry representatives, two from industry associations, two from trucking companies, and one from a major Australian truck insurer about the usability of the questionnaire. All of these representatives advised that the questionnaire was too long, and that some of the questions were repetitive. Also financial performance (of companies) was originally included, but the industry experts advised that companies might be reticent to provide accurate answers about this because they might not want this to be revealed. Based on the outcomes of these test results the questionnaire was refined, and the number of questions was reduced to 44.

The survey (Appendix B) was developed to assess whether the participating companies used a range of safety management practices identified in the literature. Although not exhaustive, the characteristics that

87 were included were judged to cover the main areas of safety management practice. The questionnaire, primarily in forced choice format, took approximately one hour to complete as an interview and covered the following broad areas. i) Descriptive information about the company – 10 questions including: the type of freight carried, the number of trucks of different size classes used, the number of employee and subcontract drivers regularly used and the lengths of trips and types of work assigned them, the number of employee drivers in different age groups (≤25, 26-55, 55-65, >65) the number of management staff involved in the transport function, annual heavy vehicle fuel consumption, and the number of recorded driver injuries in the past year. Companies recruited through telephone listings were also asked the number of vehicle insurance claims made in 2007-2009 to allow classification of their claim rate as low or higher. ii) Safety-related policies and management systems at the company – 5 questions including: whether (yes/no) policies exist on a range of individual safety-relevant issues at the company (e.g., fleet management, fitness for duty, driver training etc.), how safety policies are applied to contract drivers, whether (yes/no) drivers formally agree to comply with policies, whether (yes/no) safety is a formal performance criterion for managers, and whether (yes/no) the company participates in a range of individual safety accreditation schemes. iii) Usual vehicle acquisition, maintenance and fleet turnover practices – 7 questions including the age of vehicles at purchase, the length of truck retention, the existence (yes/no) and nature of a truck disposal policy, the average heavy vehicle truck age, the existence (yes/no) of a vehicle purchasing policy, whether (yes/no) a range of individual safety features (e.g., anti-lock brakes etc.) are included in the purchasing policy or considered in the absence of a policy, whether (yes/no) and how often trucks are scheduled for regular maintenance, and the total fleet days lost to mechanical problems in a year. iv) Work scheduling, route selection, journey planning and risk assessment – 4 questions including: whether (yes/no) consideration is given to safety in route and journey planning, whether (yes/no) a range of individual safety features (e.g., rest area availability etc.) are considered, whether (yes/no) risk assessments or safety audits are done at own depots, what proportion of delivery sites are assessed or audited for safety (all/most/some/very few/none), how driver schedules and rosters are determined (i.e., centrally, locally, using scheduling software), and whether (yes/no) a range of specific strategies are used to monitor schedules. vi) Driver recruitment – 3 questions including: who undertakes recruitment, whether (yes/no) and which particular qualifications are required, and which particular history and performance checks (e.g., employment references, licence currency etc.) are routinely conducted before hiring a driver. vii) Driver pay arrangements – 2 questions including: the basis of pay for driving (e.g., hourly rate, rate per km etc.) and whether (yes/no) non-driving activities such as loading/unloading and queuing/waiting are paid.

viii) Safety training and education – 5 questions including: whether (yes/no) a range of individual types of training are provided to drivers, which of a range of communication strategies are used to inform drivers about safety matters, whether (yes/no) and how schedulers are trained in fatigue management, whether (yes/no) and what type of safety training is provided to managers, and whether (yes/no) safety managers have undertaken any Health and Safety training. ix) Safety management participation arrangements – 2 questions including: whether (yes/no) drivers can participate in safety management and what specific mechanisms (e.g., Health and Safety committee etc.) they can use, and what process is applied to deal with safety concerns raised by drivers. x) Driver behaviour management – 5 questions including: whether (yes/no) individual forms of in-vehicle performance monitoring are used, what actions are taken in response to working hours breaches and other unsafe behaviours, whether (yes/no) and what system is used to analyse safety incidents, and whether (yes/no) and what positive incentives are provided for safe performance.

4.2.4 Procedure In the first round of recruitment, the insurance policy numbers of companies meeting the claim rate and selection criteria were extracted from the de-identified claims data of a major insurer. The criteria were chosen to maximise the differences between company size groups and between claims groups while ensuring roughly equal numbers of candidates in the four categories. The insurer contacted the insurance broker for each of these companies, explained the purpose of the study and requested the broker seek permission from the company for researchers to contact them. When permission was received, the thesis author or research assistant (helping the thesis author) contacted the company to invite a representative to take part in the survey. Unfortunately many insurance brokers were unwilling to invite their clients to participate, resulting in very few invitations and, consequently few acceptances. After persevering with this recruitment strategy for some time, this method was abandoned due to a lack of cooperation by brokers to agree to the thesis author contacting their clients.

Instead, companies were recruited to the study through invitations to a random selection of transport companies listed in the Yellow pages. In the second round of recruitment, the thesis author telephoned transport companies with publicly listed phone numbers directly to explain the study and invite a representative to participate.

Information about the study was sent to all interested companies and formal consent was obtained from willing participants. All participants were provided with a copy of the survey questions in advance. For both methods of recruitment, to maximise convenience and thus participation, volunteer companies could choose to complete the survey as a telephone interview at a mutually convenient time, an online survey or a paper survey returned via mail or scanned email. All participants were given a $75 gift voucher to compensate their time. The study was approved by a UNSW Human Research Ethics Advisory Panel (Approval number: 08/2011/52).

89 4.2.5 Response rate As indicated in Table 4. 1, the overall response was 14% of those candidates who were contacted. While this is much lower than the intended response, studies of response rates for surveys in the trucking industry confirmed similar response rates using similar approaches to that found in the current study (Lau, 1995; Lawson and Riis, 2001).

A total of 84 managers volunteered to complete a survey with .a higher response rate for those contacted directly from the randomised list of companies in the Yellow Pages than for those identified based on their insurance policy listing. Information about insurance claim rates and company size could not be determ ined in advance for companies recruited through the telephone listing, so questions on fleet size and claims were added to the questionnaire. Applying the same recruitment eligibility criteria as applied to the insurance policyholder population meant that only 33 of the 67 companies recruited from the randomised Yellow Pages list met the eligibility criteria and were included in the final sample. These criteria were used as they were based on objective outcomes (safety-related claims) and a large population of companies.

Table 4.1. Survey response rates and number eligible for the study using criteria established from insurance criteria for those who completed the survey questionnaire using both methods of recruitment Sample source NVolunteered N Contacted Response Rate N Eligible Insurer 17 199 0.09 17 Yellow Pages 67 404 0.17 33 Total 84 603 0.14 50

The study involved a tailored design method in which study participants were allowed to select their preferred mode of response (Dillman, 2009) - telephone interview, online survey or paper survey. Mode of response was fairly evenly spread across the study groups. Of the 50 participants in the final sample, 6 lower-claiming and 15 higher-claiming companies completed an online survey, 3 lower-claiming and 7 higher-cl aiming companies completed a telephone interview, and 7 low-claiming and 12 higher-claiming companies completed a written questionnaire.

As the final sample size was smaller than anticipated, comparisons were not conducted in the final analysis between small and large companies. The selection of companies on size, however, controlled for the effect of this variable in the comparison of good and poorer performers as each group contained roughly equal proportions of small and large companies (small-low n=9, large-low n=11 and small-high n=12, large-high n=18).

4.2.6 Survey analysis method Given the small sample size, the usual chi square statistical tests would not produce statistically significant find ings for most variables being compared. Because of the small sample size, the focus was on

identifying larger effect sizes between good and poor performing companies. The importance of associations was based on effect sizes of the odds ratios and risk ratios to compare the probability of a characteristic being present in organisations with good safety track records compared with those with poorer safety track records.

Based on recent work (Olivier and Bell, 2013) on effect sizes for 2×2 contingency tables, the odds ratios were considered meaningful when they were 1.86 and above, or the reciprocal 0.54 or below. These are considered to be medium to large effect sizes based on Cohen’s (1992) effect sizes for ϕ relative to the maximum attainable correlation ϕmax. To guard against the possibility of over-interpreting results based on very low or high proportions of companies, medium effects were only considered important if the percentages of positive responses were not < 10% nor > 90% or, if they were < 10% or > 90% if the lower and higher claim group differed by more than 10 percentage points.

Thus, comparisons of binary variables for low and higher-claiming companies and mean values of continuous variables were made. However, sometimes the spread of the data was too great for using this method for mean values (for example, larger fleet sizes ranged from 14 to 886). In these cases, log-linear regression was used. The negative binomial was used in lieu of the Poisson due to over-dispersed data. Rates were analysed similarly using the rate denominator as an offset variable. Where the distribution was too skewed to obtain adequate model fit, Mann-Whitney tests were used.

4.3 Results of the manager survey (Study 2)

The 50 participating organisations came from various industry sectors, such as local government councils, utility companies, and freight transport companies. All of the companies had at least one employee driver and 25 companies (12 lower-claiming and 13 higher-claiming) employed subcontractor drivers, either on a regular basis and/or as freelancers, as well as employees.

The comparison between better and poorer performing companies revealed a number of differences. As a consequence, the results of the comparisons between organisations with low and higher claim rates were organised into the following sections according to overarching management topics: freight and vehicle fleet, journey and risk management, staffing and driver recruitment, policies and safety accreditation, scheduling and training, communication and driver participation in WHS, work monitoring, driver discipline and safety incentives, and incidents and record keeping. For parsimony, only the data on practices and characteristics showing evidence of differences between low and higher-claiming companies have been tabled and in a few cases, differences have been described only in the text.

4.3.1 Freight and vehicle fleet The types of freight carried and fuel usage per truck did not differ between low and higher-claiming companies. It was noted that the types of businesses differed in that 30% of the lower claimers were not for hire and reward transport compared with only 7% of the higher claimers.

91 Of the companies in the large company categories, those with higher claim rates tended to be larger with an average of 65 trucks compared to lower-claiming large companies that had an average of 19 trucks in their fleets. This difference in absolute fleet size was significant (rate ratio = 0.29, p = 0.001), but it should be noted that the high mean number of heavy trucks for higher claimers was inflated by a minority of companies with fleets that were noticeably larger 11 9-886 trucks than the others (greater than 60 trucks). To explore the relationship between claim rate and fleet size further, the claim rate per truck was examined for higher-claiming, large companies (with >1 4 trucks). The higher-claiming, large companies with smaller fleets (less than 60 trucks) had a lower mean cl aim rate (0.38 claims per truck, SD = 0.16; n=13) than the higher-claiming, large companies with larger fleets (0.65 claims per truck, SD = 0.39; n=5) but the difference was not statistically significant (Mann-Whitney U = 18, Standardised test statistic = 1.430, p = 0.173).

Neither truck maintenance schedules nor days off road due to mechanical failure drtferentiated lower from higher cl aimers.

The odds of higher-claiming companies were less than half those of lower claimers in considering safety features when purchasing vehicles (Table 4.2). There was some indication in the data that consideration of Electronic Stability Programs (ESP) and front or rear underrun devices also had higher (double) odds among higher claimers, but the analysis was limited by the small number of companies considering each feature in their purchase decisions. None of the companies considered rear underrun devices alone, so when they considered them, they were also likely to consider other devices as well. For this reason, consideration of front underrun, rear underrun and electronic stability program (ESP) technology were combined. The odds of companies considering any of the three safety features were greater among higher claimers than lower claimers. A similar percentage of larger (24%) and smaller (19%) companies considered any of these three devices.

Table 4.2 Considerations involved in truck purchasing decisions by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers Higher Lower claim rate claim rate Safety considerations applied to truck purchasing Odds Ratioa n (%) n (%) (N=30) (N=20) At least one vehicle safety feature considered in purchase 16 (53) 14 (70) 0.490

ESP or front or rear underrun devices 9 (30) 3 (15) 2.429 a Medium to large effect sizes were considered to be meaningfu l and were identified where odds ratios were >1.86 or <0.54 (Olivier and Bell, 201 3)

4.3.2 Journey and risk assessment Companies were asked if they consider route safety factors when planning journeys, including grade separation, overtaking lanes, bridge capacity, road conditions, rest area availability, HAZMAT routes, over-

92 dimensional vehicle access, speed limiting on poorer quality roads, traffic conditions, weigh stations, safety-cams, tunnels and low underpasses. There were few differences in responses by higher and low claimers, but the organisations that had low claim rates more often checked traffic conditions when planning transport journeys and considered speed limiting on poorer quality roads than did those with higher cl aim rates (Table 4.3). Companies wer·e also asked if they audit their own worksites and the delivery sites. The odds ratios show that lower-claiming companies conduct risk or safety assessments of their own sites compared with higher-claiming companies.

Table 4.3 Journey and site risk assessment by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than low claimers. Higher Lower claim rate claim rate Journey and site risk assessment practices Odds Ratioa n (%) n (%) (N=30) (N=20) Check traffic conditions prior to journeys 6 (20) 8 (40) 0.375 Speed limiting on poorer quality roads 3 (10) 4 (20) 0.444

Carry out safety audits at own worksites 24 (80) 19 (95) 0.211 a Medium to large effect sizes were considered to be meaningful and were identified where odds ratios were >1 .86 or <0.54 (Olivier and Bell, 2013)

4.3.3 Staffing and driver recruitment There was no significant difference between higher and low claimers in their use of sub-contractor drivers, nor their transport manager to driver ratios.

The companies were asked about aspects of driver histories and performance that were checked prior to employment. Examples of recruitment checking criteria that were given in the questionnaire were: references from previous employers, licence currency, licence points, accident history, in-vehicle driving performance, selection test performance, health or other. Those with lower claim rates more often checked accident histories of drivers before hiring them (Table 4.4). On the other hand, the odds for lower claiming companies to check references, licence points or conduct performance tests when recruiting drivers were less than for higher claimers. In addition, companies were asked how many employee drivers were under 25 years of age, 26-55 years of age, 55-65 years of age and over 65 years of age. The higher claiming company odds ratio for employing any drivers over the age of 65 was more than four times that of lower-claiming companies, and higher claimers had nearly twice the odds of employing drivers aged 55-65, but showed no differences in the younger age groupings. Table 4.4 shows the results for staffing and recruitment practices with medium or greater effect sizes.

93 Table 4.4 Driver staffing and recruitment practices for companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers Higher Lower claim rate claim rate Odds Driver recruitment and staffing practices n (%) n (%) Ratioa (N=30) (N=20) Use recruitment firm 6 (20) 2 (10) 2.25 Check accident history 18 (60) 15 (75) 0.500 Check references 28 (93) 16 (80) 3.500 Check licence points 20 (67) 8 (40) 3.000 Conduct performance test 6 (20) 2 (10) 2.250 Has employee drivers over age 65 9 (32) 2 (10) 4.263 Has employee drivers between ages 55 and 65 19 (66) 10 (50) 1.900 a Medium to large effect sizes were considered to be meaningful and were identified where odds ratios were >1.86 or <0.54 (Olivier and Bell, 201 3)

Higher and lower claimers did not differ on a number of aspects of the recruitment process including whether subcontractors were responsible for recruiting drivers. Higher claimers had higher odds of using a recruitment firm for hiring, but this method of recruiting drivers was relatively uncommon.

4.3.4 Pay and conditions Companies were asked how drivers (employees, subcontractors and freelance drivers) were paid for driving. The options included hourly rate, flat day rate, day rate with overtime, weekly rate, weekly rate with overtime, salary, flat rate per truckload, trip rate (based on kilometres travelled or tonnage carried) or other method. Compared to lower-claiming companies, the odds of higher claimers paying employees by productivity (trip or truckload) methods were nearly five times higher (Table 4.5). All of the lower-claiming companies utilised time-based pay, i.e. hourly or salaries, for at least some of their employee drivers, compared to 87% of higher claimers. It should be noted that 3 lower-claiming and 3 higher-claiming companies used time-based and trip-based payment methods to pay their employee drivers, most probably reflecting different freight tasks assigned to different employee drivers within these companies.

There was no difference between lower- and higher-claiming companies in the ways that employee drivers were remunerated for loading and unloading. However, employee drivers in higher-claiming companies had lower chances of being paid for hours spent queuing o:r waiting than employee drivers in lower claiming companies.

94 Table 4.5 Driver payment practices for companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers Higher Lower Odds Driver payment methods claim rate claim rate Ratioa n (%) n (%) For Employees N=30 N=20 Employees paid for time worked 26 (87) 20 (1 00) - b Employee drivers paid either per truckload or trip 14 (47) 3 (15) 4.958 Employees paid for all hours spent queuing or waiting 24 (80) 19 (95) 0.211 For Subcontractors (for companies that employ them) N=8 N=4 Subcontractor drivers paid a trip rate 3 (38) 3 (75) 0.200 Subcontractor drivers paid for all hours spent queuing or 7 (86) 2 (50) 7.000 waiting For Freelance drivers (for companies that employ them) N=10 N=8 Freelance drivers paid a trip rate 6 (60) 2 (25) 4.500 Freelance drivers paid for all hours loading and unloading 2 (20) 4 (50) 0.250 a Medium to large effect sizes were considered to be meaningful and were identified where odds ratios were >1 .86 or <0.54 (Olivier and Bell, 2013) b Odds ratio could not be calculated as there was no variation in one group.

When subcontractor drivers were employed, the odds of higher claimers paying them trip rates were less than the odds of lower claimers and the odds of paying for queuing and waiting time were higher. On the other hand, higher claimers were more likely than lower claimers to pay freelance drivers a trip rate and the odds of freelance drivers being paid for all hours loading or unloading were lower- at higher-claiming companies than lower-claiming companies.

4.3.5 Policies and safety accreditation The survey asked about a number of safety policies including: WHS risk assessment, WHS audits, WHS reporting systems, fleet management, driver recruitment and selection, driver training, fitness for duty, fatigue management, on-road behaviour, depot behaviour, work monitoring, work planning (e.g. journey planning), driver performance monitoring, seat belt use, vehicl e selection, vehicle maintenance, depot conditions, accident prevention and response. The analysis showed that the odds of higher claimers having policies on fleet management, fatigue management and work monitoring were considerably greater than for lower claimers (see Table 4.6). No differences were found between the claim groups for any of the other policies.

With regard to safety management accreditation programs, the survey asked if the company was accredited under the National Heavy Vehicle Accreditation Scheme (NHVAS) Mass Management, NHVAS Maintenance, NHVAS Basic Fatigue Management, NHVAS Advanced Fatigue Management, TruckSafe, ISO 9001 or other schemes. Higher claimers showed greater odds of being accredited under the NHVAS Mass Management and NHVAS Basic Fatigue Management compared with the lower cl aimers. No other

95 important differences were found for the claim groups on membership of accreditation programs. Respondents were also asked if managers in their company had heavy vehicle safety management key perform ance indicators (KPis). The odds for higher claimers having these KPis were nearly three times greater compared with those for lower claimers.

Table 4.6 Safety policies and safety accreditation by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers

Higher Lower claim rate claim rate Safety policies and accreditations Odds Ratioa n (%) n (%) (N=30) (N=20) Safety policies on fleet management 20 (67) 8 (40) 3.000 Safety policies on fatigue management 25 (83) 10 (50) 5.000 Safety policies on work monitoring 14 (47) 5 (25) 2.625 NHVAS Mass Management 11 (79) 3 (21) 3.281 NHVAS Basic Fatigue Management 11 (37) 4 (20) 2.316 Have key performance indicators for safety management 10 (35) 3 (15) 2.982 a Medium to large effect sizes were considered to be meaningful and were identified where odds ratios were >1.86 or <0.54 (Olivier and Bell, 201 3)

4.3.6 Scheduling and training Respondents were asked whether drivers' schedules are determ ined centrally, whether local depots schedule and roster drivers, whether loads are scheduled centrally and rosters determined locally, whether rostering/scheduling software is used or if they managed scheduling and rostering in some other way. Companies with lower claim rates showed higher odds of scheduling work centra lly, whereas higher claiming companies had greater odds of scheduling loads centrally while rostering drivers loca lly and/or doing both locally (Table 4.7). Asked about whether schedulers were trained in fatigue risk management, a total of 20 higher claimers and 14 lower claimers answered this question with 15 higher claimers and only 6 lower claimers saying that they are trained, so the odds of higher claimers doing this was 4 times greater than lower claimers. The survey asked what type of safety training was provided for drivers, including the following types: WHS induction, Workplace Health and Safety, fatigue risk management, driving skills (on road}, driving skills (classroom), pre-trip inspection, manual handling, loading/unloading, or other. The odds of higher-claiming companies reporting conducting various types of safety related training for their employee drivers, including fatigue risk management, driving skills (on road and in classroom}, eco driving and pre-trip vehicle inspection, were greater compared to the lower-claiming companies.

96 Table 4.7 Scheduling and training by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers Higher Lower claim rate claim rate Odds Scheduling and training practices n (%) n (%) Ratioa (N=30) (N=20) Schedules are determ ined centrally 12 (40) 13 (65) 0.359 Loads are scheduled centrally and rosters determined locally 7 (23) 1 (5) 5.783 Local depots schedule and roster drivers 10 (33) 4 (20) 2.000 Fatigue risk management training for employee drivers 21 (70) 10 (50) 2.333

Driving skills (on road) training for employees 17 (57) 5 (25) 3.923 Driving skills (classroom) training for employees 7 (23) 1 (5) 5.783 Eco driving (fuel economy) training for employees 5 (17) 1 (5) 3.800 Pre-trip vehicle inspection training for employees 14 (47) 6 (30) 2.042 a Medium to large effect sizes were considered to be meaningful and were identified where odds ratios were >1.86 or <0.54 (Olivier and Bell, 2013)

The survey asked whether managers responsible for WHS were provided specific safety training. No differences in this practice distinguished between higher and lower claimers.

4.3.7 Communication and driver participation in workplace health & safety The survey asked how safety inform ation was communicated to drivers, in particular asking whether they used newsletters, noticeboards, supervisors, union representatives, toolbox talks or staff meetings, special briefings, text messages, email, staff website or other means. There was little difference in methods or mean number of communication methods used by companies with lower claim rates and by higher claimers, both averaging three modes of communication. The survey also asked whether and how drivers are involved in WHS decision-making - e.g. through a WHS committee or union representative, toolbox or staff meetings, suggestion box or other means. The odds of opportunities for driver input to safety decision-making were less in companies with higher claims than in companies with lower claims (Table 4.8). Driver involvement via a WHS committee was more likely among higher than lower-cl aiming companies. When asked if there was a time limit for addressing drivers' safety concerns, the odds of higher claimers were less than those of lower claimers to set and monitor time limits for responding to drivers' safety concerns.

97 Table 4.8 Communication and driver input for companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers Higher Lower claim rate claim rate Odds Methods of safety communication and driver participation n (%) n (%) Ratioa (N=30) (N=20) Drivers are involved in WHS decision-making 24 (80) 19 (95) 0.211 Drivers are involved via a WHS committee 14 (47) 6 (30) 2.042 There is a time limit for dealing with drivers' safety concern s 11 (38) 11 (61) 0.389 The time limit is monitored and enforced 9 (82) 11 (100) _ b a Medium to large effect sizes were considered to be meaningful and were identified where odds ratios were >1.86 or <0.54 (Olivier and Bell, 201 3) b Odds ratio could not be calculated as there was no variation in one group.

4.3.8 Work monitoring Respondents were asked about monitoring driver hours and schedules. They were asked to indicate if hours and schedules are monitored, and how they were mornitored. The choices were: in-truck logs or work records are reviewed after an incident, random spot checks of in-truck logs or work diaries are carried out, all in-truck record ings are reviewed, all diary recordls are reviewed or hours and schedules are monitored in another way. None of the work monitoring variables distinguished between higher claimers versus lower claimers.

The survey also asked whether the companies used other fomns of in-vehicle monitoring, including: GPS tracking of the vehicle, fuel consumption, braking analysis, gear change analysis, speed analysis, fatigue monitoring systems (e.g. Optalert) or other devices. Higher-claiming companies were more likely to report GPS tracking, fu el consumption monitors, and speed analysis devices. Lower claimers responded with a range of other types of monitoring practices not listed in the questionnaire choices, such as sending new drivers out with experienced drivers, documenting fatigue management records, and pre-start checks. Table 4.9 contains the findings about monitoring driver behaviour.

Table 4.9 In-vehicle monitoring by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers Higher Lower claim rate claim rate Odds Type of in-vehicle monitoring used n (%) n (%) Ratioa (N=30) (N=20) Use GPS tracking device In-vehicle monitoring 19 (63) 7 (35) 3.200 Use fu el consumption In-vehicle monitoring 13 (43) 4 (20) 3.059 Use speed analysis In-vehicle monitoring 9 (30) 3 (15) 2.429 Use other in-vehicle monitoring systems 1 (3) 4 (20) 0.138 a Medium to large effect sizes were considered to be meaningful and were identified where odds ratios were >1.86 or <0.54 (Olivier and Bell, 201 3)

98 4.3.9 Driver discipline and incentives Higher and lower claimers were found to differ on some questions about the company's procedures for dealing with breaches of working hours and the actions taken on unsafe behaviour. Higher claimers had greater odds of investigating the reasons for breaches of working hours and of applying discipline or penalties for breaches. In contrast, higher claimers had lesser odds of using other methods for dealing with breaches of working hours, of doing formal investigations when people behave unsafely and of using other actions when people behave unsafely (such as issuing a letter to the driver, preparing notes for safety meeting discussions, or in smaller companies, directly counselling the driver or dismissing them) compared to lower claimers (see Table 4. 10). Higher claimers had lower odds of using incentives to encourage safety innovations than did lower claimers.

Table 4.1 0 Driver discipline practices and safety incentives by companies with higher and lower insurance claim rates where higher odds ratio indicates higher claimers have larger odds than lower claimers Higher Lower claim rate claim rate Odds Use of driver discipline practices and safety incentives n (%) n (%) Ratioa (N=30) (N=20) Formally discipline and penalise drivers for breaches of working 8 (27) 2 (10) 3.273 hours Investigate reasons for breaches of working hours 14 (47) 5 (25) 2.625 Use other methods to deal with breaches of working hours 2 (7) 4 (20) 0.286 Formal investigation when people behave unsafely 6 (20) 7 (35) 0.464 Other actions when people behave unsafely 3 (10) 4 (20) 0.444 Incentives are provided for safe work or management safety 5 (17) 7 (35) 0.387 innovations a Medium to large effect sizes were considered to be meaningful and were identified where odds ratios were >1.86 or <0.54 (Olivier and Bell, 2013)

4.3.10 Incidents and record keeping No significant differences were found between higher and lower-claiming companies in terms of the records they keep on incidents, injuries, infring:ements or (truck) defect notices, nor in regards to the infringement rates per employee drivers. The incidence of defect notices differed between higher and lower-claiming companies with lower-cl aiming companies having a mean rate of 0.10 and median of 0.03 defect notices per truck versus 0.34 and 0.15 defect notices per truck respectively for higher-claiming companies (p = 0.02).

While the data did not reveal significant differences in lost time injuries per driver between higher and lower claimers, lower claimers reported a trend for fewer average non-lost time injuries per driver (0.03, n=18) than did higher claimers (0.07, n=27; Mann-Whitney U = 173, Standardised Test Statistic = -1.86, p = 0.06). Importantly, this finding confirms that the variable used to index safety performance, a company's

99 2007-2009 claim rate, is consistent with at least one important measure of safety outcome record ed at the time the survey was conducted in 2011 -2012.

4.4 Discu ssion of the manager survey findings

The results of this study showed some safety-related characteristics that distinguished companies with higher and lower safety performance outcomes. Some of these characteristics were as would be expected, but the survey also produced some surprising results. These results are summarised in Table 4. 11 . Overall, better perform ing (lower-cl aiming) companies were smaller, with fewer trucks. They also had fewer defect notices, did more safety-related checking and monitoring such as site safety audits, checking traffic conditions, speed limiting on poorer quality roads, and checking accident history at recruitment, were more likely to pay employee and freelance drivers for all time worked, actively monitored driver work and work load, and paid active attention to policy and compliance by having a formal approach to policy breaches, seeking driver input into WHS and responding quickly to safety concerns.

Table 4.11 Summary of expected and unexpected findings comparing 37 safety characteristics of low and higher insurance claimers where higher odds ratio indicates higher claimers have larger odds than lower claimers Survey findings of items (17) more likely Survey findings of items (20) more likely to to be prevalent in lower claimers than in be prevalent in higher claimers than in higher claimers (as expected) lower claimers (not expected) Fleet Fewer defect notices Lar er fleets Safety features in choosing vehicles New vehicles with underrun, ESC Scheduling Schedule and roster centrally Risk assessment Check traffic conditions Speed limitinq on poorer quality roads Safety au dits at own sites Recruitment/ Check accident history Check references, license points, use selection employment tests Fewer drivers over 65 and 55-65 Pay/conditions Do not pay by trip/load Less likely to pay subcontractors a trip rate Pay drivers to wait Policies Policies on fatigue risk management Policies on fleet manaqement Policies on work monitoring Accreditations Accredited NHVAS Basic Fatique Manaqement and KPis Accredited NHVAS Mass Management Have safety KPis Training Safety training - fatigue risk management Drivinq skills traininq - on road Drivino skills trainino - classroom Drivinq skills traininq - eco-drivinq Pre-trio inspection trainino Communication/ Encourage driver input into WHS driver input Faster response to safety concerns Monitoring Document fatigue management In-vehicle monitoring devices: GPS

Experienced drivers check/coach drivers In-vehicle monitorinq devices: fuel Pre-start checks In-vehicle monitorino devices: soeed Discipline/ Formal investiqation of unsafe behaviour Formally discipline drivers for hours breaches incentives Offer incentives for safety innovations Investigate reasons for hours breaches

100

Some of the characteristics that the literature would suggest should be more prevalent in HV companies with lower insurance claim rates than higher claim rates were not found. Counterintuitive findings, particularly in the areas of driver safety training, safety policies, and in-vehicle monitoring are difficult to explain. It appears that companies with higher claim rates conduct more safety related training than do the lower-claiming companies. This finding could mean that training is not one of the most important safety management practices, the type, or method, of training provided is not effective, or training was introduced in response to safety incidents and so has not yet had a positive effect on safety. This effect may be at least partly related to smaller company size where training is less important because more direct and interactive control of work practices have more effect.

Higher-claiming companies also had a greater number of safety policies than did lower-claiming companies. The number of safety policies may not, however, reflect the level of the safety management in these companies. The existence of policies does not necessarily imply that they have been accepted by all of management nor been adopted into the ‘way we do things around here’ (Hopkins, 2006; Zohar, 1980). This issue, dubbed “paper compliance” (Quinlan et al., 2010), also applies to membership of accreditation systems where their simple existence cannot necessarily be an indicator that they have been adopted throughout the company. Further research and more in-depth examination is needed to determine whether these are just documents or they are actively promoted and enforced, and whether these were introduced as a response to a problem with (claim) incidents. Lower-claiming companies seem to have a more formal and systematic approach to dealing with policy breaches. This may suggest that while rules exist in companies with higher claim rates, they may not be backed up as much with an active or formal safety management approach. Again, company size may also be influential as policies may have a more direct effect on middle managers and drivers in small companies with fewer layers and less dispersion of influence compared to large companies with many layers of authority and communication.

Moreover, while the higher claimers have greater odds of being accredited under the NHVAS Basic Fatigue Management scheme than lower claimers, the literature contains mixed results on the value of accreditation schemes. One American study found that the safety performance of HV companies improved after certification to the ISO 9000 standard (Naveh and Marcus, 2007) and another found positive safety outcomes associated with ISO 18001 (Abad et al., 2013), but the reports of safety performance linked with NHVAS are inconclusive (Mooren and Grzebieta, 2011). Clearly, more research is needed to understand the role that accreditation systems play in encouraging safer performance in the heavy trucking industry.

Higher claimers were more likely to have policies on fleet management, work monitoring and fatigue management, as well as provide more training on fatigue, driving and pre-trip inspections. They were also more likely to be accredited under an NHVAS scheme as well as having key performance indicators for safety. It may be that accreditation requires policies, training and key performance indicators. This may explain why accredited companies are more likely to have these policies, KPIs and to offer training. Higher claimers may be more likely to want accreditation under these schemes for the purposes of gaining

101 concessions that make their operations more efficient. It is also noted that higher-claiming companies in the study were far more likely to be for hire and reward transport companies and hence more interested in transport operational efficiencies than those companies that use heavy trucks only to transport their own goods and equipment.

Again, contrary to expectations, higher claimers used more methods of pre-employment checks when recruiting drivers, including reference and licence checks and (driving) performance tests. On the other hand, lower claimers more often checked accident history prior to employing drivers. This may suggest that checking accident history is a more effective selection criterion than other pre-employment checks. Lower-claiming companies less often than higher claimers employed drivers over the age of 65 or drivers between the ages 55-65. The observed relationship between older employee drivers and claim rates also needs further investigation to interpret this finding.

In work monitoring, lower claimers more often used less technological forms of monitoring such as sending new drivers out with experienced drivers and doing pre-start checks, but higher claimers used a greater number and range of in-vehicle monitoring systems than did lower claimers. It may be that these companies have particular operational needs for installation of truck monitoring devices and meeting audit requirements for carrying greater mass and working longer hours, or for tracking delivery times. How and why monitoring devices are used – and whether or not the monitoring devices are checked in real time or the audit requirements are maintained between audits is unknown.

There is a much greater tendency for higher-claiming companies to pay employee drivers by truckload or per trip, i.e., on a piece rate basis, compared with lower-claiming companies who tend to pay drivers on a time basis, whether on an hourly or weekly pay basis or by salary. The by-trip or truckload payment method has been found in other research to be linked with poor safety performance (Quinlan and Wright, 2008b; Rodriguez et al., 2006; Williamson, 2007; Williamson and Friswell, 2013). Therefore, this finding is consistent with the expected results based on previous research. When lower claimers employed subcontractors, they paid them by trip but were more likely to include payment for queuing and waiting. Both higher and lower claimers used trip pay for some employee drivers, which may reflect different freight tasks assigned to different employee drivers within these companies.

Like employee drivers freelance drivers were paid by trip and not paid for any other aspects of work at higher-claiming companies, whereas most lower-claiming companies employing freelance drivers paid for all hours of work performed. This pattern of findings reinforces prior research findings that paying drivers for all hours of work performed is associated with better safety outcomes.

However, surprisingly the survey found that higher claimers were more likely to pay their subcontracted drivers for hours worked than were lower claimers who paid subcontractors by trip or truckload. It may be that, especially in for hire and reward transport companies, in higher-claiming companies there is a greater tendency for regular drivers to be engaged via subcontractors rather than direct employment contracts

102

(while using the same payment method as they would employee drivers) compared with lower-claiming companies, that may tend to hire drivers on employment contracts if they regularly use them.

Higher-claiming companies reported the use of more types of communication modes with drivers. This could imply that the quality and type of communication is more important that the number of modes of communication used. However, the quality of communication was not tested in this survey. As expected, driver participation in WHS decision-making was more prevalent in companies with lower claim rates compared with higher claim rates. This is consistent with the findings in 16 studies reported in Chapter 3.

The general impression from the survey results is that the heavy vehicle transport companies with lower insurance claim rates per truck tend to take a more active and substantive approach to managing safety in their organisations, whereas the higher-claiming companies take a more passive or static method of managing safety. The higher claimers tend to rely more on setting criteria and rules for vehicles and drivers, than do the lower claimers. Lower claimers seem to focus more strongly on proactive risk assessment, ensuring that rules are agreed, and consulting drivers on safety issues, adding more strength to prior findings that vigilant risk assessment (Banks, 2008; Saksvik et al., 2003), communications (Lu and Tsai, 2010; Wills et al., 2005) and effective relations between drivers and managers (Al-Refaie, 2013; Kath et al., 2010) are predictors of safety outcomes.

4.4.1 Limitations of Study 2 Clearly, the company response rate is a limitation of this study. A considerably higher response rate was intended, however, low response rates are a continuing problem with this study population. A number of other studies involving light and heavy truck drivers in Australia (Friswell, 2013; Williamson et al., 2009) and internationally (Peignier et al., 2011) have had similar response rates. Furthermore, two studies of response rates for surveys in the trucking industry confirmed similar response rates using similar approaches to that found in the current study (Lau, 1995; Lawson and Riis, 2001). It is likely that the margins are so small and pressures so great that managers of companies do not have the time to respond. The low response rate is symptomatic of an industry under pressure. Nevertheless, this study took a unique approach to understanding the predictors of safe performance in this industry by actively comparing the characteristics of companies with demonstrated good and poorer safety outcomes. The results therefore provide the foundation for further research to show whether the characteristics that distinguished good and poorer performing companies in this study and should be confirmed in another study.

It had been hoped that by recruiting trucking companies through an insurer would have provided a sample of candidates with validated safety outcome data, and that better response rates would have been achieved. Problems were encountered, as the insurance brokers, who were the link with companies, were unwilling to allow the thesis author to contact companies directly despite strong assurances that any participation in this study would be anonymous. Consequently, the calculated response rates almost

103 certainly underestimated the actual response rates, as many brokers (but unfortunately an unknown number) probably did not invite participation on behalf of the thesis author.

Given the low numbers of participants recruited for this study, performance differences between large and small companies were not pursued separately in the analysis. Although fleet size was controlled by ensuring the two performance groups had similar representation of larger and smaller companies, fleet size was associated with claim rates since higher claimers had on average a larger number of trucks in the fleet. This effect clearly needs to be pursued in further research. In addition, as vehicle kilometres travelled was not taken into account, some differences between higher and lower-claiming companies may have been a function of more exposure to on-road risk. However, the finding that fuel usage per truck (an index of distance travelled) was similar in lower and higher-claiming companies argues against this possibility. Further research is needed to understand whether this has an effect on safe performance.

Another limitation was that the claims status was established for a period before the survey period with the claims data from 2007-2009, and survey period in 2012. Therefore, the safety management features reported in 2012 may have been put in place as a response to claims in the prior 3-year period, or there may have been other significant changes such as company acquisitions. However, the non-lost time injury comparisons showed that higher claimers still tended to have poorer safety performance than lower claimers in 2012 despite any management efforts implemented after 2009.

Finally, the cross-sectional nature of the study may have masked possible explanations of seemingly counter intuitive findings. For example, companies with higher claim rates may have introduced more policies, training and monitoring systems as a response to their heightened insurance claims. However, this study was not able to determine when or if these changes occurred before, during or after the period selected for comparison.

4.5 Conclusions of the manager survey (Study 2)

The value of this study is that it showed that some safety management characteristics distinguished heavy vehicle transport companies with lower insurance claim rates from with those with higher claim rates. While there may be important characteristics or combinations of characteristics that could not be tested in this study, this research provides some directions for further research by indicating where to look further to examine aspects of safety management that can result in better safety and crash outcomes.

Clearly, pay methods are important considerations in safety management. All of the lower-claiming companies paid their employee drivers for the time worked, whereas the higher claimers had nearly five times the odds of paying drivers by truckload or per trip by comparison.

The unexpected results, such as higher-claiming companies having more safety policies, assist to refine the focus of further research into the content and implementation of these policies. Those with lower insurance claims reported having more active management characteristics than higher claimers, such as checking traffic conditions, speed limiting on poorer quality roads, conducting safety audits, gaining WHS

104

ideas from drivers, acting on drivers’ safety concerns, and taking a formal approach to addressing safety breaches.

Conversely, the higher claimers reported having safety equipment and policies, but it was unclear, from the survey, how these were used in practice. It may be that applying specific criteria in the selection of drivers, trucks and monitoring equipment, together with having a range of safety policies is not as important as other practices such as risk assessments, enforcement of rules, and involving drivers in safety management.

The study has provided evidence on effectiveness of some characteristics of companies that, potentially if implemented in companies that are poorer performing, in terms of crash outcomes, and that do not already have these characteristics, will improve safety practices and outcomes. However, the many counterintuitive findings, suggest that more research is needed to provide a richer evidence base for understanding the important safety management characteristics.

It should be noted also, that this survey data was self-reported safety management information from company representatives. The information that they reported may not be accurate, either deliberately or because they really do not know what goes on in their company, or because they want to appear better than they are. For these reasons, the findings of the study need to be validated. This is especially important as the aim of the project is to develop a method or intervention that should improve safety through confirming the factors that make a difference in enhancing safety, and with these, building a safety management system.

Some of the results of the good-poor safety characteristic comparisons were not as expected, so a validation of the existence of reported characteristics was needed to ensure that the proposed contents of the SMS are likely to be backed by strong evidence.

The next stage of this research (Study 3), involved in-depth investigations in order to validate, and further explain the features found in this survey that distinguish better safety performing companies and that could form the basis of an effective safety management system. It was considered important to validate the self- reported characteristics by collecting observable data and re-interviewing managers and to look at the consistency between managers and drivers in reports about characteristics regarding their work sites.

This further research, reported in Chapter 5, should help to better define how elements of safety management are applied and received in organisational settings. Moreover, observations of combinations of safety management characteristics in situ should reveal a fuller picture of how a safety management system should be constructed.

105 Chapter 5: In-depth Investigation Study (Study 3)

This chapter discusses the third study undertaken in the thesis research. Study 3 was an in-depth investigation of a sample of Study 2 participants for the purpose of validating the original survey findings.

5.1. Introduction

The manager survey (Study 2), discussed in Chapter 4, provided evidence of characteristics that distinguish better and poorer performing companies in terms of safety management. Findings of the survey pointed to important distinguishing characteristics in the management of vehicles, drivers and work environments.

5.1.1 Specific Study 2 findings The characteristics found in the original survey to distinguish between lower and higher claimers can be grouped into those that are about managing the risk environment, managing drivers and managing the safety climate of the company. These are briefly summarised below.

Characteristics to manage the risk environment The trucks are the immediate work environments of truck drivers. With regard to managing vehicles, the survey found that lower insurance claimers were more likely to consider at least one safety feature when purchasing trucks. These companies also had fewer instances of receiving defect notices; and of the companies with large fleets, those with very large numbers of trucks had higher claim rates per truck than the large companies with lower claim rates. However, contrary to what the literature suggests (Cantor et al., 2010; de Pont, 2005; Wills et al., 2005), there was no detectable difference in maintenance frequencies between higher and lower claimers. Lower claimers were more likely to consider at least one feature when purchasing trucks, but higher claimers more often considered electronic stability programs (ESP) and front or rear underrun devices for their trucks. Moreover, higher claimers were more likely to be accredited under the NHVAS Mass Management Scheme. With regard to managing the broader work (risk) environment, higher claimers were more likely to have safety policies on fleet management, fatigue management and work monitoring, contrary to expected results. They were also more likely to be accredited under NHVAS Basic Fatigue Management (BFM). Lower claimers were more likely to schedule delivery journeys centrally, check traffic conditions and limit speeds on poorer quality roads than did the higher claimers. Moreover, lower claimers were more likely to carry out safety audits at their own work sites than were higher claimers and to be responsive to drivers’ safety concerns..

Characteristics to manage risk associated with managing drivers Findings about the management of drivers showed a number of important differences between lower and higher claimers. A distinguishing factor was the method of payment to drivers, where lower claimers were more likely to pay drivers for time worked, whereas higher claimers more often paid a trip or truckload rate. Lower claimers more often checked accident histories of drivers when recruiting, but higher claimers more

106

often checked references, license points and used a test for driving performance. Contrary to expectations (based on the research literature (Gander et al., 2005; Gregersen et al., 1996)), higher claimers were more likely to provide safety training, including fatigue risk management, driving skills and pre-trip inspection. They were also more likely to use in-vehicle monitoring devices including GPS, speed and fuel consumption monitoring. With regard to disciplinary practices, higher claimers were more likely to investigate reasons for, and apply formal disciplinary actions for breaches of working hours. But consistent with expectations, lower claimers were more likely to apply formal disciplinary procedures for unsafe behaviour.

Characteristics to manage risk and safety climate While there was no difference between lower and high claimers in the types and numbers of communication modes used, lower claimers were more likely to provide opportunities for driver input into WHS decision-making.

5.1.2 Case for validating Study 2 survey findings (Study 3) There were expected and unexpected findings across all aspects of safety management in the findings of this survey. Most of the important safety management characteristics found to be more prevalent in lower- claiming companies were what would be expected based on prior safety management research. However, the presence of a number of unexpected findings suggested there was a need to validate the findings and to develop a richer understanding of the practices at the companies and how these might be influenced by company characteristics and contexts.

Social desirability bias (Grimm, 2010) may have come into play in this survey (Study 2). As this was a self- report survey completed by managers about their own safety management practices, there is risk that they might have answered the questions in a way that would show their practices in a more positive light than may have been the case (Wåhlberg et al., 2010). That is, for example, their responses may reflect more about what they think they should be doing instead of what they actually were doing. Moreover, the managers may not have known the answers to some of the questions and just guessed the answers.

This study tested the validity of managers’ survey responses to the characteristics included in the survey described in Study 2 by conducting in-depth interviews with a sample of managers who participated in the original survey. The in-depth interviews provided the opportunity to determine whether managers are consistent in their responses to the same questions they were asked in Study 2. In addition, the validity of managers’ responses was also assessed by surveying a small number of drivers from each of the companies participating in the in-depth interviews. These surveys provided converging information about the validity of managers’ initial responses in the survey. Driver responses were not treated as strong validation for all characteristics examined in Study 2 because it is possible that driver responses may not have reflected the true situation in the surveyed companies, or they may not have known some of the answers. For example, they possibly did not know the recruitment criteria applied in driver selection

107 processes. The driver survey therefore did not include these characteristics. Lastly, the in-depth study attempted to collect documentary and observational data to support or refute managers’ claims where possible. While some managers had provided documented evidence of safety policies, pre-trip inspection forms and other documents, at the time of the survey, to support what they had claimed, most did not.

In summary, therefore, a qualitative in-depth study was conducted with a sample of surveyed companies to validate or refute the survey findings. This study also sought to explain the context of findings of the survey. While qualitative research methods carry some risk of pre-conceptions shaping the analysis (Malterud, 2001), care was taken to minimise researcher bias, by focusing on specific verifiable characteristics identified in the scientific literature, by comparing manager self-reported characteristics with driver experience and perceptions, and finally by on-site observation. The rationale for this study, Study 3, is based on two premises:

1. Study 2 was a survey involving interviews and self-reported safety management practices and there may be reasons for the information provided to be more favourable than reality, and

2. The study was limited to what could be asked using such a methodology and although the interview/survey was based on the findings of previous research (Study 1) about the factors that should be considered, it was still somewhat limited in scope.

For these reasons, it was important to do a further in-depth study to validate the findings of Study 2. The aims of this additional study were to:

a) validate Study 2 manager survey findings using verbal and visual data collected from managers in Study 3; b) interpret the effects of lack of consistency to confirm the validity of conclusions from Study 2 regarding the differences in lower- and higher-claiming companies’ safety management characteristics; and c) examine the nature of the differences where drivers’ interview responses were inconsistent with managers’ survey responses.

5.2. Methods

5.2.1 Design The approach taken in this study was similar to an audit of management practices aiming to understand the management systems in place, examine safety instruments used, and gather and evaluate evidence of their applications (Brommelsiek and Tinsley, 1996). Practices were audited against the information provided in the earlier survey of companies targeted for this study. In other words, the aim was to investigate the accuracy of the reported safety management practices by looking at these practices in more detail in a small sample of the companies that participated in the original survey, similar to the approach adopted by Banks (2008) in her study of light vehicle fleet safety management. Beyond this, the

108

aim was to clarify the nature and context of these safety management characteristics. A mixed methods approach was used. Observational data, organisational documentation and interview data from managers and drivers were compared to the original survey responses.

The design of the study was to enable gathering of evidence to confirm management practices reported in the original survey. Managers interviewed were asked more probing questions about these practices as well as for documented or visual evidence that these practices were in place. Driver interviews were conducted to provide evidence of safety management practices independent of managers’ reports.

5.2.2 Participants Study 3 was an in-depth investigation of Australian organisations that operate heavy trucks, and that were identified as higher- and lower- safety performers. The originally surveyed 50 companies were stratified by size and claim rate category (large lower-claiming, large higher-claiming, small lower-claiming and small higher-claiming) and then randomly ordered. The companies were then contacted in the four groupings in this order until four companies in each group were recruited or the list was exhausted. The aim was to recruit 16 companies in total.

The managers in the participating companies were asked to assist the recruitment of three drivers in their company to participate in confidential interviews. Because drivers generally spend little time at the company site, convenience sampling was the method of recruitment. That is, while managers were urged to use a random selection process (alphabetise surnames and select every second driver), in practice, the companies provided drivers who could be available on the day and time of the site visit.

Two of the 50 potential candidate companies (Study 2 participants) had gone out of business at the time of recruiting candidates for Study 3. The remaining 48 companies were put into four groups – small lower claimers (n=9), small higher claimers (n=12), larger lower claimers (n=11), and large higher claimers (n=17). Then, the companies within each group were randomised. Starting at the top of each list, companies were invited to participate. The aim was to include 4 companies from each group – large lower- claiming, large higher-claiming, small lower-claiming and small higher-claiming companies. However, only three of the 11 companies in the large lower-claiming category agreed to participate in the study. For the remaining three groups four companies were recruited as intended, although in one small lower-claiming company, no drivers agreed to participate. This meant that because only four of the seven companies in this group agreed to participate, it was not possible to recruit another company that may have provided three drivers to interview. Further, in one small higher-claiming company only one driver completed the survey. Note that drivers were offered a number of options for the interview methods, including completing the interview by phone, at a stopping location convenient for them, or completing the questions in writing. In all, 38 companies were approached to participate and 15 agreed to participate in the study, a participation rate of 39.4%. Similar proportions of the Study 3 sample were for-hire-and-reward companies. In Study 2 two thirds (66.7%) of lower claimers and 93% of higher claimers were for-hire-and-reward

109 transport companies, and in Study 3, 57.2% of lower-claiming companies and all higher-claiming companies were for-hire-and-reward heavy vehicle operators. As the differences in proportions of lower- and higher-claiming companies were similar, the validation research results were unlikely to be less robust.

A $100-voucher, to use at any Coles Myer store, was given to each participating company to compensate the managers’ time provided to assist the in-depth research investigations. Each participating driver was given a $50-voucher to use at any Coles Myer store. The study was approved by a UNSW Human Research Ethics Advisory Panel (Approval number: 9/13/027).

5.2.3 Measures and procedures Companies were contacted first by email outlining the intended purpose of the research investigation and what the research team would need from the participants. Companies that agreed to participate were visited at a mutually convenient time. The researcher conducted interviews with the managers of companies who completed a questionnaire in the earlier survey, collected or sited documentary evidence and made site observations at company depots and headquarters.

In two cases, both higher claimers, the companies had undergone some changes and different individuals had been appointed to senior manager positions. In these cases, the researcher also interviewed other managers who had been with the company at the time of the survey. A data-recording sheet (Appendix C) was used to guide a detailed interview with a senior manager or owner of each participating company as well as collection of supporting documents and observations.

This sheet started with questions to identify any changes in management practices since the manager completed the original survey. Following these questions, a number of questions were asked about how trucks, drivers and safety risks were managed in the company. While these questions were similar to those used in the prior questionnaire to enable confirmation of information previously provided by the managers, the method also enabled the collection of more evidence for the answers provided by the managers. The questions asked of managers were not as detailed as the survey questionnaire to avoid leading the managers to the answers previously provided.

In addition, a survey (Appendix D) based on the questions asked of the managers in the first survey was used to determine consistency of drivers’ perceptions of the safety management practices with the managers’ reported practices. The questions asked of drivers were framed to ask about their own personal experience and observations. That is, for example, drivers were asked questions about the trucks that they normally drive, rather than trucks in the fleet generally. Drivers were not asked about some items included in Study 2 where they were unlikely to know the answers. For example, drivers were not asked about pre-employment recruitment checks.

For most drivers the survey was by interview but in three of eight higher-claiming companies, drivers completed the survey in writing. In addition to examining the characteristics that were reported in the survey (Chapter 4), the in-depth interviews with managers and drivers included a question about how

110 comprehensive the safety management system is for the company. Table 5.1 provides a description of the data that was sought through the in-depth investigation process.

Table 5.1 Types of data collected in the original survey and in the in-depth investigations Survey data Manaqer interview Drivers' interview Documentation Observations Freight carried Changes to freight Freight carried by Photographic Visual evidence of carried each driver evidence freight Truck numbers and Changes to fleet Type of truck normally Records of trucks Visual evidence of types driven owned/operated truck types Truckages Truck disposal criteria Age of truck normally - - asked driven Safety features Safety features Safety featu res on Records of trucks Visual evidence of considered in considered truck normally driven safety features purchasina Maintenance Maintenance Drivers' assessment Records of truck Visual evidence of frea uencies freauencies of maintenance maintenance conditions of trucks Defect notices Defect notices Defects on truck Records of defect - normally driven notices Days off road due to Days off road due to How often does the Records of days off - breakdowns breakdowns truck breakdown road Recruitment check Recruitment check - Recruitment policy - types and criteria types and criteria Ages of drivers Ages of drivers Age of each driver - - Remuneration Remunerations Each driver's - - methods used methods used remuneration method Incentives for safety Incentives for safety Incentives for safety - - innovations innovations innovations Safety policies in the Safety policies in the Safety policies in the Safety policies - company company company Safety related Safety related Safety related Accreditation - accreditations accreditations accreditations certificates Communication Communication Communication Documented Posters, mailboxes, methods used methods and practices methods and practices communications verbal exchanges etc. Driver input methods Driver input methods Driver input methods Documented driver Verbal exchanges available and practices and practices input Safety training offered Safety training offered Safety training offered Records of training - courses Fatigue risk training Fatigue risk and Fitness and fatigue Policies, training - fitness policies, fatigue risk records assessmenUtrainina traini na Scheduling practice Driving risk and Scheduling practice Schedules, driver - and trip lengths fatigue risk and trip lengths diaries (loumey risk) assessments Site risk assessments Site risk assessments Site risk assessments Audit procedures Safety observations at site In-vehicle fatigue risk Driving risk Hours and schedule Daily run sheets - monitoring (monitoring assessment monitoring hours) Other in-vehicle How driving behaviour In-vehicle, in-vehicle - Observation of how behaviour monitoring is monitored monitoring on driver's and what is monitored truck Discipline for Discipline for Procedure for Fatigue management - breaches of working breaches of working breaches of working policies hours hours hours Discipline for unsafe Discipline for Action taken when Disciplinary policies - behaviour breaches of safety people !behave policies unsafelv System for Procedure for Incident investigation Incident and risk - investigating incidents responding to process reporting forms incidents Comprehensiveness Managers' Drivers' assessment Work procedures, Observations of ofSMS assessment of safety of safety management codes of conduct managers' and management drivers' attitudes

111 The purpose of using the expanded interview tool with managers, combined with observations and interviews with drivers about their experiences, was to cross validate information from all sources including the responses to the original survey. Manager and driver interview data, documents and observational data were collected to examine consistency with previously self-reported company safety management characteristics in the original survey (Study 2). The evidence of safety management features reported in the earlier self-reported survey, and how these features are implemented in the organisation were recorded. The evidence collected that did not validate what was reported in the earlier survey was also documented.

Lastly, the in-depth interview included questions about the manager’s perceptions of how comprehensive the safety management system is in their company, especially relating to degree of commitment to the priority of safety, characteristics of leadership and responsibility for safety.

5.2.4 Analysis method To assess the validity of the earlier survey, the manager and driver interview data, documents and observational data were compared to the manager survey results collected in the previous study. Three sets of comparisons were made. First, the survey questionnaire responses by managers during the original survey were compared for consistency of findings with the evidence from manager interviews in the current study. Where available, documents and other visual evidence were examined. For each safety management characteristic studied, the comparison was rated as consistent, non-consistent, partially consistent or changes to practice. For example, if a manager had reported having specific safety policies but did not provide consistent answers about these policies at the site visit, this characteristic was rated as inconsistent. Second, other observed or documented information related to safety management practices provided additional evidence to validate or refute the earlier survey responses, again using the same ratings of consistency. Finally, survey responses from three drivers in each company were compared against the earlier manager survey responses and the same ratings made for consistency. Where three drivers were interviewed, the characteristic tested was deemed consistent if at least two drivers gave answers consistent with their managers’ survey answers. In this way, the interpretations aimed to clarify the strength and nature of validated Study 2 findings.

In addition, where there was a reported change in management practices or safety management characteristics this was recorded but treated as partially consistent or inconsistent with earlier reported practices. In this regard, where a manager reported, for example, that he/she was making greater use of in-vehicle monitoring information for safety counselling, as distinct from introducing new in-vehicle monitoring equipment, this was rated as partially consistent.

For each management characteristic studied, frequencies of companies with evidence of full consistency with survey results were compared within size and claim rate groupings. Where there was evidence of partial consistency or inconsistency, the nature and contexts of these differences were further examined to

112

better understand the safety management characteristics distinguishing lower- and higher-claiming companies, especially whether the inconsistent responses threatened the validity of the conclusions drawn from the Study 2 survey of managers.

Thus, the validity of managers’ responses in the Study 2 survey was tested by comparing each manager’s responses on each characteristic with evidence from their in-depth interview as well as visual observations and documentary evidence available in their company. Where these indicators were found to be inconsistent, the nature of the differences in responses from Study 2 and Study 3 was examined to determine whether the inconsistency threatened or confirmed the original conclusion regarding differences between lower- and higher-claiming companies. Consistency between managers’ initial survey responses and drivers’ responses were examined separately as inconsistencies between manager and driver responses provided a different insight into circumstances in each company. The nature of any inconsistency between manager and driver was examined again to determine whether it threatened the validity of the conclusions regarding differences between lower- and higher-claiming companies.

5.3 Results of the in-depth investigation (Study 3)

The results are organised for each group of characteristics of the safety management system assessed in Study 2. For each group of characteristics comparative consistency of manager and driver responses and observational evidence are examined first for the size/claim rated company groups. For these comparisons the companies included in the tables as numerators are only those that showed clear and complete consistency with the earlier survey responses. Documents and observations were not always possible to obtain. Therefore documentary or observational evidence was treated as supplementary evidence to support validation.

Then the in-depth evidence for each characteristic was examined further to determine whether or not the inconsistent evidence from manager and observational evidence represented a threat to the validity of the conclusions drawn in Study 2 regarding the differences between lower- and higher-claiming companies. Where evidence confirmed only half or fewer than half of company survey responses within lower and higher-claiming groups, these characteristics were considered as not conclusively confirmed as characteristics that distinguished between lower and higher claimers.

Lastly, inconsistencies between manager responses and driver responses were scrutinised to interpret the extent and nature of the inconsistency. Inconsistent responses where managers changed their response from one that supported the conclusion drawn from Study 2 about a particular characteristic to a response that did not support the conclusion, cast doubt about the validity of the conclusion. In contrast, if the inconsistency involved changing from a response that did not support the conclusion to one that did, this would support the Study 2 conclusion. In both cases, the extent of inconsistency depends on the proportion of managers changing their response, where a larger proportion provides greater support in either direction.

113 5.3.1 Truck management characteristics Proportions of companies showing Study 2 and Study 3 consistency about trucks In Table 5.2 the numbers of companies in each of the four size/claim rated groups that provided evidence of consistency with earlier survey responses is presented for the group of characteristics relating to truck management characteristics including: size of fleet, defect notices and safety features on trucks. For each of these characteristics, managers from lower-claiming companies gave consistent responses with their original survey responses, about how they manage the trucks, between survey and in-depth interviews. By contrast, those managers from higher-claiming companies were much less consistent regarding defect notices and on whether they consider safety features. Drivers were asked only about safety features, including electronic stability programs (ESP) and underrun protection devices. Again driver responses from lower claimers were more consistent with their managers than those from higher claimers. The availability of documentary evidence was relatively low for most characteristics, and for both lower and higher claimers.

5.3.1.1 Manager-provided Study 3 findings about trucks Further examination of the implications of consistent and inconsistent evidence from manager-provided data between Study 2 and Study 3 regarding truck fleet management is shown in Table 5.3. All managers interviewed were consistent in reporting the size of fleet in their company so validating the conclusions of Study 2. Regarding numbers of defect notices only two higher claimer managers were inconsistent with one indicating higher numbers of defect notices and one indicating lower numbers of defect notices, thus validating the original finding that lower claimers had fewer than higher claimers.

Higher claimer managers were also inconsistent in reporting whether their company considered safety features in truck-purchasing decisions as 62.5% of managers in higher-claiming companies advised that they do not consider safety features when purchasing new trucks contrary to their responses in the survey. As this finding is consistent with the conclusion from Study 2, it is validated.

5.3.1.2 Driver survey findings about trucks Driver survey findings about consistency with the original managers survey (Study 2) relating to safety features in new trucks and ESP and underrun protection devices in trucks are shown in Table 5.4. Drivers reported what safety features they actually had fitted to the trucks that they normally drove. With regard to safety features, all drivers in lower-claiming companies reported having the same features on their trucks as were indicated as truck purchasing decision-making criteria in the Study 2 survey. Drivers in only 25% of higher-claiming companies reported having truck safety features that were reported in the survey by managers in making truck purchasing decisions so while this contradicted their managers’ responses, it was consistent with the conclusion from Study 2. Just over one third of drivers from both lower and higher claimers contradicted their manager responses on having ESP and/or underrun protection devices on their trucks and in both cases, these responses were inconsistent with the conclusion drawn from Study 2. While it is possible that drivers did not know what features are fitted to their trucks, the nature of this

114

inconsistency does not support the original survey finding that higher claimers are more likely to consider ESP and/or underrun protection devices when purchasing trucks. The level of inconsistency is insufficient to invalidate the survey finding, but no conclusion on these features can be confirmed from driver responses.

Summary of Study 3 results about trucks In summary, truck related safety management characteristics found in the survey to distinguish between lower and higher-claiming companies, and validated by consistent responses by managers in this study, include truck numbers, the relative number of defect notices and consideration of at least one safety feature considered in purchasing new vehicles. Driver responses also confirmed the conclusions regarding safety features in purchasing policies whereas driver evidence was inconclusive on ESP and underrun protection devices.

115 Table 5.2 Consistency for truck management characteristics- numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent.

...... """J In-depth validation (Study 3) (Study 2 Results)

Lower- versus Higher Large lower-claiming a Large higher-claiming a Small lower-claiming a Small higher-claiming a claiming companies (n=3) (n=4)c (n=4) b (n=4)

Finding V> c V> c V> c .'1 -1!? Q) Q) - 0 -1!? Q) Q) - 0 -1!? Q) Q) -0 -I!? (see Table 4.2 on Q) u -~ (,) c,.,. Q) u ...f!? (,) c,.,. Q) u ...f!? 0 c,.,. c Q) ~ -!!? ~ Q) c ~ ro ~ ro ~ ro Q) Characteristic ~!5 > :;:, 2: ~!5 %! !5 :;:, 2: ~!5 %! !5 :;:, 2: ~c %! c :;:,~ c~ ·c~ c~ · c~ c~ ·c ~ c Q) ·c Q) ro > Oii') 8 3l ro > Oii') 8 3l ro > Oii') 8 3l 0:2 8 :::;ea> :::;ea> :::iE Q) ~~ > o-8 o-8 o-8 Q) Q) 0 Truck numbers 3/3 4/4 4/4 4/4 Defect notices 3/3 1/3 4/4 4/4 Safety features (any) 3/3 3/3 3/4 2/4 4/4 3/3 1/4 0/4 0/4 ESP or front or rear 2/3 3/4 2/3 2/4 underrun protection • ·-· is not reported due to data not collected. b No drivers completed the survey for one Small Lower-claiming company. c Two of these companies had undergone significant changes in their safety management between the time of the survey and the in-depth investigation.

11 6 Table 5.3 Manager-provided evidence regarding trucks - percentages of companies showing consistency with their responses in the original survey, through managers' interview responses and observations/documentation.

Characteristic %Manager Nature of inconsistency Conclusion consistent Truck numbers 100.0 Validated

Defect notices 100.0 71.4 One Higher claimer reported fewer defects, one Validated reported more defect notices. One Higher claimer had not answered the and was excluded. Safety featu res 100.0 37.5 Five Higher claimer managers advised that they did Validated not consider safety features contrary to their survey

ESP or Managers were not Managers were not asked. No underrun asked. conclusion

Table 5.4 Driver-provided evidence regarding trucks- percentages of companies showing consistency with their managers' responses in the original survey through drivers' interview responses.

Characteristic % Drivers Nature of inconsistency % Drivers Nature of inconsistency Conclusion consistent consistent Truck numbers Drivers were not asked. Drivers were not asked. No conclusion Defect notices Drivers were not asked. Drivers were not asked No conclusion Safety 100.0 25.0 Drivers in six Higher-claiming companies Consistent featu res did not report having featu res on their trucks that their managers had claimed to consider when trucks. ESP or 66.7 Drivers in two Lower claimers reported 62.5 Drivers in three Higher claimers said that No underrun that their trucks did have one or more of their trucks did not have these features conclusion these features contrary to their contrary to their manager's survey

117 5.3.2 Scheduling, journey and site risk assessment and response to drivers’ safety concerns In Table 5.5 the numbers of companies in each of the four size/claim rated groups that provided evidence of consistency with earlier survey responses is presented for the group of characteristics relating to risk assessment and management. These included: scheduling, journey and site assessment and time limits on managers’ responses to drivers’ safety concerns. Managers were not interviewed about how scheduling was organised nor time limits on dealing with safety concerns, but they showed consistent responses for journey risk assessments involving traffic and speed limits on poorer roads as well as site risk assessment for both lower and higher claimers. Drivers were asked about central scheduling practices, and time limits on dealing with safety concerns. Drivers’ evidence on scheduling was inconsistent with findings of managers’ responses in the original survey for both claimer groups but was mostly consistent for timeliness of responses to safety concerns with the exception of large higher claiming companies.

5.3.2.1 Manager-provided Study 3 findings about scheduling, journey and site risk assessment Table 5.6 shows the extent and nature of constancy with the original survey data, based on evidence from managers’ interviews in lower-claiming and higher-claiming companies, with respect to journey and site risk assessments. Managers were not asked if their scheduling and rostering were done centrally or locally nor about timeliness of dealing with safety concerns. With regard to journey assessment, the managers’ interview responses confirmed the survey finding that lower claimers were more likely than higher claimers to check traffic conditions and to limit speeds travelled on poorer quality roads so validating the Study 2 conclusion. There was documentary or observational evidence for these practices in two lower-claiming companies and none in higher-claiming companies.

Similarly, for site risk assessments, most managers’ responses were consistent and the patterns of inconsistencies for lower claimers confirmed that they carry out site risk assessments and that higher claimers did not. In three out of seven lower-claiming companies copies site risk assessment forms were provided, compared with only one in eight higher claimers. One manager in a higher-claiming company said in the survey that site risk assessments were conducted, but when interviewed, he said that “drivers are meant to do these, but they don't.” Therefore, this characteristic was validated.

5.3.2.2 .Driver survey findings about scheduling and time limits for dealing with drivers’ safety concerns Driver interview results with respect to scheduling, as well as time limits for responding to drivers’ safety concerns, are provided in Table 5.7. A notable percentage of drivers reported scheduling and rostering practices that were inconsistent with their managers’ responses in Study 2 and with the conclusion from Study 2 that lower claimers scheduled centrally. In driver interviews, the responses suggested that scheduling was more of a team effort in lower-claiming companies compared with higher-claiming companies. Drivers in one lower-claiming company advised that the most important safety practice in their

118

company was meeting each morning and planning the workday together. By contrast, drivers in a higher- claiming company advised that rosters prepared by local managers were alright, but scheduling of loads was done by another part of the company and that there were often lengthy delays in loading, resulting in drivers starting night delivery journeys sometimes three hours after the start of their shift. As a result of these inconsistent findings, no conclusion could be made about central versus local scheduling and rostering.

The original survey found that lower-claiming companies were more likely to report setting time limits for responding to driver concerns about safety matters than higher claimers. One lower claimer and no higher claimers provided a procedure document indicating time limits for managers’ responses to safety concerns raised by drivers. The drivers in three lower-claiming companies said that although there were no time limits, contrary to their managers’ survey responses, they advised that time limits were unnecessary as managers were very responsive whenever a safety concern was raised. Whether or not the drivers knew if the managers had formal time limits, the interpretation of this finding is that it supports the Study 2 finding that lower claimers are more likely than higher claimers to set time limits. Conversely, in half of higher- claiming companies who had reported in the survey that there were time limits for acting on drivers’ safety concerns, drivers interviewed in the study indicated there were no time limits or they did not know if there were. These drivers did not indicate that limits were not needed. The overall driver survey finding confirms the original survey finding.

5.3.2.3 Summary of findings about scheduling, journey and site risk assessments and time limits for dealing with drivers’ safety concerns In summary, the finding from manager responses that lower claimers were more likely to schedule delivery journeys centrally was not tested in the interviews with managers and the responses by drivers calls the conclusion from Study 2 into question. However, the driver data suggests that lower claimers are more likely than higher claimers to report a greater consideration in ensuring schedules and rosters that would not compromise safety. The findings that journey and site risk assessments are more likely to be carried out in lower-claiming companies than in higher-claiming companies were confirmed by the study. In addition, the finding that lower claimers were more likely than higher claimers to have time limits or pay attention to time on responding to drivers’ safety concerns was confirmed by driver data.

119 Table 5.5 Consistency for scheduling, risk assessment and responding to drivers' concerns - numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent. In-depth (Study 3)

Large lower-claiming a Large higher-claiming a Small lower-claiming a Small higher-claiming a (n=3) (n=4)c (n=4)b (n=4)

V> c V> c V> c .'1 Q) -o -~ Q) Q) - 0 -~ Q) Q) -0 -~ -1!? <.J c,.,. Q) 0 -~ 0 c,.,. Q) 0 ...f!? 0 c,.,. c Q) ~ -~ ~ Characteristic Q) c ~ nl Q) c ~ nl ~ nl Q) Q) > :;:, 2: :ti'~ > :;:, 2: :ti'~ %! ~ :;:, 2: :ti'c %! c :;:,~ ·c::: :"2 c:-2 ·c::: :"2 c:-2 ·c::: :"2 c Q) ·c::: Q) oa; 8 $ nl > oa; 8 $ nl > oa; 8 $ 0:2 ::::?!Q) ::::?! Q) ~~ > 8 0~ 0~ 0~ Q) Q) 0 Schedules centrally' 2/3 2/4 1/3 1/4 Journey risk"" - check 3/3 2/3 3/4 3/4 3/4 traffic conditions/ Journey risk.. - speed 3/3 4/4 4/4 4/4 limiting on poorer roads Site risk assessment.. 3/3 2/3 3/4 1/4 3/4 1/4 3/4 3/3 1/3 1/4 2/3 3/4

120 Table 5.6 Manager-provided evidence regarding scheduling, risk assessment and responding to drivers' concerns - percentages of companies showing consistency with their responses in the original Survey through managers' interview responses and observations/documentation.

Characteristic % Manager Nature of inconsistency % Manager Nature of inconsistency Conclusion consistent consistent Managers were not asked. Managers were not asked No conclusion Journey risk 85.7 The one inconsistent Lower-claiming 75.0 Both Higher claimers, who were inconsistent, Validated assessment - company manager had indicated in the reported that they did not do these in check traffic survey that checking traffic conditions contradiction to their survey answers. conditions was not undertaken, but demonstrated in this that were undertaken. Journey risk 100.0 100.0 Validated assessment -

85.7 The manager in one Lower-claiming 75.0 Both Higher claimers who were inconsistent Validated assessment company, who had reported in the reported that they did not do these in survey that site risk assessments were contradiction to their survey answers. not done, advised in interview that they are done. Managers were not asked. Managers were not asked. No conclusion

121 Table 5.7 Driver-provided evidence regard ing scheduling, risk assessment and responding to drivers' concerns - percentages of companies showing consistency with the responses by their managers in the original Survey through drivers' interview responses.

Characteristic Finding from Study 2 % Drivers Nature of inconsistency % Drivers Nature of inconsistency Conclusion survey consistent consistent Schedules Lower were more likely to 50.0 Drivers in one Lower-claiming company 37.5 Drivers in half of the Higher-claiming No centrally do than Higher advised central scheduling contrary to companies said that their companies conclusion the survey. In two Lower claimers scheduled and rostered centrally contrary to drivers said scheduling was done their managers' survey response, and locally, when the manager survey another company driver said scheduling was indicated central scheduli done local their Journey Lower were more likely to Drivers were not asked. No do than Higher Site risk Lower were more likely to Drivers were not asked. Drivers were not asked. No assessment do than Higher conclusion Time limit for Lower were more likely to 83.3 Drivers in one Lower-claiming company 50.0 Drivers in half of the Higher-claiming Consistent dealing with do than Higher said there were time limits inconsistent companies said there were not time limits safety concerns with their managers' survey and drivers. contrary to what was reported in the survey. The nature of the inconsistencies supports the su

122

5.3.3 Driver employment and remuneration There was good consistency between managers’ original survey and in-depth interview responses for all driver remuneration characteristics but only for managers from lower-claiming companies (see Table 5.8). For higher-claiming companies, manager responses were relatively consistent about employing older drivers and payment types, but were inconsistent for details of recruitment practices. Little documentary or observational evidence was available to validate these characteristics as the recruitment procedural documents obtained did not specify driver selection criteria. Drivers from both lower- and higher-claiming companies largely supported the responses of managers in the original survey with respect to their reported ages and remuneration. But they were not asked about driver recruitment criteria because drivers are not likely to know about the driver recruitment checks undertaken.

5.3.3.1 Manager-provided Study 3 findings about driver employment and remuneration As shown in Table 5.9, while all lower claimer managers gave consistent responses to questions about checking accident histories when recruiting drivers, all but one of the higher-claiming companies’ managers provided inconsistent responses about the driver recruitment checks they reported in the survey (Study 2). Interestingly, despite reporting in the survey that they did not check accident histories, two higher claimers showed job application forms that ask driver recruits to record their accident histories. Most of the inconsistent responses from higher claimer managers involved changing their response to admit that they did not conduct licence points checks, nor test drivers for safety performance. Also, the three managers in higher-claiming companies who were consistent in saying that they check referees, the in-depth interviews showed that the checks were more focused on work attitudes than on safety behaviours. As so many high claimer managers changed their response to admit that they didn’t do reference and other checks, it questions the validity of the Study 2 conclusion. Therefore the survey finding that higher claimers more often than lower claimers carry out these safety checks was invalidated.

The finding that higher claimers were more likely than lower claimers to employ drivers over the age of 60 was confirmed by manager interviews for both lower and higher claimers. One large higher claimer showed a policy about not discriminating on the basis of age.

With regard to driver pay, there was overall good consistency between what managers in both lower- and higher-claiming companies reported in the Study 2 survey and the evidence provided by managers in the Study 3 in-depth interviews about how drivers are being paid and what they are paid for. Therefore, the findings that lower-claiming companies are more likely to pay drivers hourly pay and for all hours worked, were validated. The survey finding that lower claimers were less likely to pay subcontractor drivers for hours of work was not tested.

5.3.3.2 Driver survey findings about employment and remuneration As shown in Table 5.10, based on drivers’ actual ages reported in Study 3, the finding that higher claimers were more likely to employ drivers over the age of 60 was not tested. There was good consistency of

123 drivers responses compared with the Study 2 managers’ survey about how they were paid and what they were paid for, adding additional weight to the validity of findings with respect to remuneration for employee drivers. However, there were too few subcontractor drivers interviewed to support or refute the finding that lower claimers were less likely to pay by time worked compared with higher claimers.

5.3.3.3 Summary finding about driver employment and remuneration In summary, based on evidence provided by managers in Study 3, the finding that lower-claiming companies more often than higher claimers check accident histories when recruiting new drivers was validated. The finding that higher claimers were more likely to check references, licence points and conduct performance tests of driver recruits than lower claimers was not validated. The finding that higher claimers were more likely to employ drivers over the age of 60 was validated. The survey finding that remuneration distinguishes between lower and higher claimers was confirmed, with lower claimers more likely to pay drivers on the basis of the hours they work – whether driving or waiting for loads – compared with higher claimers. Driver responses also supported this conclusion. The Study 2 finding that lower claimers were less likely to pay subcontractor drivers for hours of work could not be confirmed with drivers because few subcontractors participated in the study.

124 Table 5.8 Consistency for employment and remuneration characteristics- numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent. In-depth validation (Study 3)

Large lower-claiming a Large higher-claiming a Small lower-claiming a Small higher-claiming a

(n =3) (n =4)c (n=4)b {n=4)

rn c rn c -c,.,.0 - Q) -ac,.,. - Q) ~ (.') ~ (.') nl Q) c nl Q) c Characteristic ~ Q) ~ Q) ::I 2: > ::I 2: > 8 $ ·co :'2a; 8 $ ·co :'2a; Cl ~ Cl~ Recruitment - check accident 3/3 0/4 2/4 4/4 1/4 history Recruitment - check references, 3/3 0/4 3/4 1/4 licence points, test perfonnance Employ drivers over 60 years old 3/3 4/4 1/4 4/4 4/4 Drivers paid per truckload or trip 3/3 3/3 3/4 3/4 4/4 3/3 3/4 3/4 Subcontractors paid for time Drivers paid for waiting 3/3 3/3 3/4 3/4 4/4 3/3 4/4 3/4

125 Table 5.9 Manager-provided evidence regarding employment and remuneration -percentages of companies showing consistency with their responses in the original Survey through managers' interview responses and observations/documentation. Lower claimers Higher claimers Characteristic % Manager Nature of inconsistency % Manager Nature of inconsistency Conclusion consistent consistent Recruitment 100.0 12.5 Two Higher claimers said they did check, Validated check accident contrary to their survey response and five said history they didn't check, contrary to their survey

Recruitment 85.7 One Lower contradicted his survey 12.5 Seven managers in Higher-claiming Not check references, answer saying he did these checks, companies said they do not do these checks validated licence points, test despite not reporting these checks despite saying they did in the survey. performance in the Employ drivers 100.0 100.0 Validated over 60 old Drivers paid per 100.0 75.0 Managers in two Higher claimers advised Validated truckload or trip changed payment methods - one per kilometre to hourly and one from hourly to per kilometre rates. Subcontractors Managers were not specifically Managers were not specifically asked about No paid for time asked about pay methods for pay methods for subcontractors. conclusion subcontractors. Drivers paid for 100.0 87.5 A manager in one Higher-claiming company Validated waiting who reported in the survey that drivers were paid for waiting contradicted this in interview. Otherwise all others were consistent.

126 Table 5.10 Driver-provided evidence regarding employment and remuneration- percentages of companies showing consistency with the responses by their managers in the original Survey through drivers' interview responses.

Characteristic Finding from Study 2 % Drivers Nature of inconsistency % Drivers Nature of inconsistency Conclusion survey consistent consistent Recru 1tment - check Lower d1d more than Drivers were not asked. Drivers were not asked. No conclusion acc1dent h1sto Higher Recru 1tment - check Lower did less than Drivers were not asked. Drivers were not asked. No conclusion references, licence Higher pomts, test erformance Employ dnvers over Lower d1d less than Drivers were not asked .. Drivers were not asked. No conclusion 60 ears old Higher I - •· • t[;J·--· Lower did less than 100.0 75.0 In two Higher claimers drivers reported Consistent I • I I Higher inconsistently to their managers about payment methods - one per kilometre to hourly and one from hourly to per

Subcontractors paid Lower did less than Insufficient numbers of No conclusion for time Higher

Drivers paid for Lower did more than 100.0 75.0 Consistent waiting Higher

127 5.3.4 Policies, accreditations, and key performance indicators (KPIs) for safety management Table 5.11 shows numbers of large and small companies that were consistent with their survey responses on safety policies, accreditations, and key performance indicators for safety management. There was good consistency of manager responses on fatigue management policies fleet management policies and for NHVAS accreditations for BFM and Mass Management but managers from large companies were not consistent on whether or not they had KPI’s set for safety. Driver responses were also largely consistent with their manager responses for all of these characteristics except drivers from higher-claiming companies on safety policies for monitoring. Documentation was available for most of these characteristics for lower-claiming companies and large, but not small, higher-claiming companies.

5.3.4.1 Manager-provided Study 3 findings about safety policies, accreditations and KPIs As shown in Table 5.12 with regard to safety policies, the finding that higher-claiming companies were more likely to have fatigue risk policies was confirmed primarily through managers’ interview responses and policy documents provided by managers. Four out of seven lower claimers and four out of eight higher claimers provided fatigue policy documents. Policies on fleet management and work monitoring were not asked of managers. However, two of seven lower claimers and no higher claimers provided these policies to view.

Managers were consistent about NHVAS accreditation, for both Basic Fatigue Management (BFM) and Mass Management, although one manager in a higher-claiming company provided an accreditation certificate during the interview - not for the scheme he had advised in the survey that his company was accredited to (TruckSafe) but rather to a different one. One large lower claimer provided a letter indicating an exemption from the Western Australia Heavy Vehicle Accreditation Scheme and two small lower claimers showed visual evidence of accreditation to NHVAS (BFM). One large and one small higher claimer showed evidence of this accreditation. One large higher claimer showed a certificate of accreditation to NHVAS Mass Management.

The survey finding that higher claimers were more likely to have key performance indicators (KPIs) for safety management was not confirmed by this study. In fact, 50% of the higher claimers (managers) said that they did not have safety KPIs, whereas in the survey they said that they did have them. In addition, three of the lower claimers who said that they did not have KPIs reported in the interviews that they did have them. Three out of seven lower claimers and only one out of eight higher claimers showed documentary evidence of safety KPI assessments. The fact that both of these results contradict the Study 2 conclusion, calls into question its validity so it was judged to be not validated.

5.3.4.2 Driver survey findings about safety policies and accreditations Table 5.13 shows that, with regard to safety policies, the findings that higher-claiming companies were more likely to have fatigue risk policies, fleet management policies and work monitoring policies were confirmed by drivers’ responses in both lower- and higher-claiming companies. With regard to NHVAS

128

BFM and NHVAS Mass Management accreditations, drivers’ interview responses supported the survey findings as well. For all of these characteristics, where drivers were inconsistent there was either no pattern of inconsistency or the driver response supported the conclusion from Study 2.

5.3.4.4 Summary of Study 3 results about safety policies, accreditations and KPIs In summary, this study confirmed the finding that lower claimers are less likely than higher claimers to have fatigue risk, fleet management, and work monitoring polices and be accredited to the NHVAS (BFM) and Mass Management. In addition, driver responses supported manager responses on all of these characteristics. The survey finding that lower claimers were less likely to have key performance indicators for safety management than were higher claimers was not validated by this study.

129 Table 5.11 Consistency for policies, accreditations and KPI characteristics- numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent.

IIIIUi:ll S UI Vt:y In-depth validation (Study 3) (Study 2)

Lower- versus Higher Large lower-claiming a Large higher-claiming a Small lower-claiming a Small higher-claiming a claiming companies (n=3) (n =4)c (n=4)b (n=4) Finding V> c V> c V> c .'1 .. -c ,.,.0 -c ,.,.0 -c ,.,.0 c (Table 4.6 on page 96) • Q) ro Q) ro Q) ro Q) Characteristic - .. E 2: E 2: E 2: E :;:, Q) :;:, Q) :;:, Q) 81:l 81:l 81:l § Cl o Cl o Cl o 0 Safety policies - fatigue Lower less than Higher 3/3 3/3 3/3 3/4 4/4 4/4 3/4 2/3 1/4 3/4 1/4 Safety policies - fleet Lower less than Higher 2/3 2/3 4/4 2/3 4/4 management• Safety policies - Lower less than Higher 2/3 214 2/3 2/4 monitoring• Accreditation -NHVAS Lower less than Higher 3/3 3/3 1/3 3/4 3/4 1/4 4/4 3/3 2/4 4/4 4/4 1/4 Basic Fatigue Management (BFM) Accreditation -NHVAS Lower less than Higher 3/3 3/3 4/4 4/4 1/4 4/4 3/3 3/4 3/4 Mass Management Safety KPis Lower less than Higher 1/3 2/3 1/4 1/4 3/4 1/4 3/4 • ·-· is not reported (due to data not oollected) b No drivers oompleted the survey for one Small Lower -claiming oompany c Two of these had in their between the time of the

130 Table 5.1 2 Manager-provided evidence regarding policies, accreditations and KPi s - percentages of companies confirming or not confirming their responses in the original Survey through managers' interview responses and observations/documentation.

Characteristic % Manager Nature of inconsistency % Manager Nature of inconsistency Conclusion consistent consistent Safety policies 85.7 One Lower claimer provided evidence 75.0 One Higher claimer said they did not have Validated - fatigue of having these policies despite these policies so contradicting their claim in the indicating in the survey that they did survey, and one said they did in contradiction to not answer. Safety policies Managers were not asked. Managers were not asked. No - fleet conclusion

Safety policies Managers were not asked. Managers were not asked. No - monitori conclusion Accreditation - 100.0 87.5 One Higher-claiming company the manager Validated NHVAS BFM provided a certificate indicating accreditation to this Scheme whereas he had reported in the survey that the company was only accredited to TruckSafe. 100.0 87.5 One Higher claimer advised that they did not Validated have it contrary to the manager's survey answer. 57.1 In three Lower-claiming companies 50.0 In fou r Higher-claiming companies managers Not validated managers who reported in Study 2 that reported that they did not have these whereas they did not have KPis demonstrated in in the survey they said they did. 3 that do have them.

131 Table 5.13 Driver-provided evidence regarding policies, accreditations and KPis - percentages of companies showing consistency with the responses by their managers in the original Survey through drivers' interview responses.

Characteristic Finding from Study 2 % Drivers Nature of inconsistency % Drivers Nature of inconsistency Conclusion survey consistent consistent Safety policies - Lower were less likely 83.3 In one Lower-claiming company drivers 62.5 In two Higher-claiming companies drivers Consistent fatigue to have than Higher claimed they did have these policies claimed having these policies when their whereas the survey claimed not to have manager had reported they did not, and in these policies. one Higher claimer drivers' evidence did not support the manager's claim to have them. Safety policies Lower were less likely 66.7 In one Lower claimer drivers said they did 100.0 Consistent - fleet to have than Higher have these policies, and in another one management driver said they did not have this policies,

Safety policies - Lower were less likely 66.7 50.0 In four Higher-claiming companies drivers Consistent monitoring to have than Higher reported that they had these policies in contradiction to their managers' survey

Accreditation - Lower were less likely 100.0 87.5. In one Higher-claiming company drivers Consistent NHVAS BFM to have than Higher reported that they had accreditation contradicti ng the survey claim that the was not accredited. Accred1tat1on - Lower were less likely 100.0 75% Drivers in two Higher-claiming companies Consistent NHVAS Mass to have than Higher reported that did not know if the company Mana ement was accredited. Safety KPis Lower were less likely Drivers were not asked Drivers were not asked. No to have than Higher conclusion

132

5.3.5 Driver input into WHS decisions and safety training As shown in Table 5.14, managers from lower-claiming companies were consistent in reporting practices regarding driver input into WHS and training for fatigue risk management and documentation was available in most companies on these characteristics. Managers from large higher-claiming companies were also consistent for both characteristics, but those from small higher claiming were only consistent in reporting fatigue risk management training practices. Driver responses were much less consistent overall with their manager’s responses. Drivers from lower claimers were consistent for driver input into WHS, fatigue risk management training and on-road driving skills training practices, but less so for reporting classroom and Eco driving skills training especially drivers from larger lower-claiming companies. Drivers from higher-claiming companies were inconsistent with their manager’s reports on almost all characteristics.

5.3.5.1 Manager-provided Study 3 findings about driver input into WHS and safety training As Table 5.15 shows for the driver input into WHS characteristic, managers in lower-claiming companies showed good consistency with the survey. Six out of seven lower claiming companies showed documentary or observational evidence that driver input into safety is encouraged. Beyond this, managers also discussed ways that they consulted drivers on truck purchases, driving plans, and schedules. Managers from higher-claiming companies were less consistent in reporting on this characteristic, but as the inconsistency involved managers who initially reported that their company allowed driver input changing their response in the in-depth interviews, the evidence for validation of this characteristic is strong. Only one in eight higher claimers showed evidence that driver input was encouraged.

The study finding on provision of fatigue risk training confirmed the finding that higher claimers provided this training more than lower claimers. Four out of seven lower claimers and three out of eight higher claimers showed documentary evidence of this training. Study 2 results on managers’ reports of the provision of driver skills and pre-trip inspections training were not tested in manager interviews.

5.3.4.2 Driver survey findings about driver input into WHS and safety training Table 5.16 shows supporting evidence from drivers from lower-claiming companies regarding opportunities for driver input into WHS decisions indicating a strong level of confidence that their managers listen and act on their suggestions. By contrast, drivers from 62% of the higher-claiming companies advised that despite what the manager indicated in the survey, they do not have input in WHS. Drivers in higher-claiming companies also expressed some doubt about whether their suggestions are taken seriously by managers. This suggests that the survey finding that lower claimers do encourage driver input into WHS decisions more than do higher claimers is validated.

Driver responses regarding fatigue risk training confirmed the finding that higher claimers were more likely than lower claimers to provide this training. Although drivers from around one-third of both lower- and higher-claiming companies gave responses inconsistent with their manager’s responses, mostly the driver

133 response was consistent with the conclusion from Study 2. For on-road driving skills training, driver interviews did not confirm the finding that higher claimers were more likely than lower claimers to provide this training. Only a minority of drivers from higher-claiming companies gave the same response as their manager on this characteristic, and all inconsistent responses did not support the Study 2 conclusion. Driver responses on classroom-based driver training were quite inconsistent and there was no clear pattern of inconsistent responses. For eco-driving training, drivers were also inconsistent compared to manager’s responses, but mostly drivers’ responses supported the conclusion from Study 2 that higher- claiming companies were more likely to use this training. Pre-trip inspection training was not tested.

5.3.4.4 Summary of Study 3 results about driver input into WHS and safety training In summary, the Study 2 finding that lower claimers were more likely to encourage driver input into WHS decision-making was validated as a distinguishing characteristic through both manager and driver responses. The finding that fatigue risk management training is less likely to be provided in lower-claiming companies was also validated by both manager and driver data. Managers were not interviewed about the three different types of driving skills training, but driver responses only concurred with their manager reports of eco-driving training, and contradicted manager responses on on-road driving skills and classroom driving skills training.

134 Table 5.14 Consistency for driver input and safety training characteristics- numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent. In-depth (Study 3)

Large lower-claiming a Large higher-claiming a Small lower-claiming a Small higher-claiming a (n=3) (n=4)c (n=4)b (n=4)

Characteristic

3/3 0/4 2/3 2/4

0/3 1/4 2/3 2/4

135 Table 5.15 Manager-provided evidence regarding driver input into WHS and safety training - percentages of companies showing consistency with their responses in the original Survey through managers' interview responses and observations/documentation.

Characteristic Finding from Study 2 survey % Manager Nature of inconsistencies % Manager Conclusion consistent consistent Driver input Lower were more likely to have 100.0 62.5 Managers in three Higher-claiming Validated into WHS than Higher companies who had said in the survey that drivers had input advised and/or demonstrated that drivers do not have · Safety training Lower were less likely to have 71.4 A manager in one Lower claimer 75.0 A manager in one Higher claimer said that Validated - fatigue risk than Higher claimed that they did provide it, and they did provide it, and another said they did management another claimed they did not, not, contrary to their survey answers. ""n' ""''~' to their su answers. Dnvmg sk1 lls Lower were less likely to have Managers were not asked. Managers were not asked. No trammg - on than Higher conclusion road Driving ski lls Lower were less likely to have Managers were not asked. Managers were not asked. No training than Higher conclusion classroom Driving ski lls - Lower were less likely to have Managers were not asked. Managers were not asked. No Eco than Higher conclusion Pre-tr1p Lower were less likely to have Managers were not asked. Managers were not asked. No mspect1on than Higher conclusion tramm

136 Table 5.16 Driver-provided evidence regarding driver input into WHS and safety training - percentages of companies showing consistency with the responses by their managers in the original Survey through drivers' interview responses.

Characteristic % Drivers Nature of inconsistencies % Drivers Nature of inconsistencies Conclusion consistent consistent Driver input into 100.0 37.5 Drivers in five Higher-claiming companies Consistent WHS said that drivers did not have input whereas the su indicated that did. Safety training - 66.7 In one Lower claimer drivers claimed 62.5 In three Higher-claiming companies drivers Consistent fatigue risk that they did provide it, and in another said that the training was provided contrary management drivers claimed they did not, contrary to their managers' survey response. to answers. Driving ski lls 83.3 Drivers in one Lower-claiming 25.0 Drivers in six of the Higher- claiming Not training - on road company said that this training was companies reported that this training was consistent provided, whereas the original not provided even though the survey manager survey indicated that it indicated that it was.

Driving ski lls 33.3 Drivers in three of the Lower -claiming 37.5 In five Higher-claiming companies drivers Not training - companies advised that this training said this training was not provided whereas consistent classroom was not provided in contradiction to the survey indicated that it was provided. their managers' survey response. Driver in one Lower claimer said that it was provided contrary to the

Driving ski lls - 33.3 Drivers in three of the Lower-claiming 25.0 In five of Higher-claiming companies, Consistent Eco companies advised that this training drivers said that the company provided this was not provided in contradiction to training whereas the survey indicated that it their managers' survey response. wasn't provided. In one Higher-claiming Driver in one Lower claimer said that company drivers said the training wasn't it was provided contrary to the provided contrary to the survey.

Drivers were not asked. Drivers were not asked. No conclusion

137 5.3.6 Monitoring, discipline and incentives As shown in Table 5.17, good overall consistency was found between survey and interview responses from managers of both higher and lower claimers for most monitoring, discipline and incentives characteristics. Insufficient data was obtained from managers or drivers to test the results about discipline and investigation for working hours limit breaches. Drivers from lower-claiming companies made responses consistent with their manager’s response for almost all of the characteristics shown. There was considerably less consistency between driver and manager responses for higher-claiming companies for most characteristics. Large lower claimers also provided the most documentary evidence compared to the other companies.

5.3.6.1 Manager-provided Study 3 findings about monitoring, discipline and incentives Table 5.18 shows that managers’ interview responses were consistent with their survey responses for characteristics including conducting pre-start checks, using experienced drivers to train new drivers and documenting fatigue so validating the Study 2 conclusion that lower claimers were more likely to use these practices than higher claimers. Four of seven lower claimers and two of eight higher claimers provided pre-start inspection forms. Interview and survey responses were also consistent or largely consistent for use of GPS tracking, validating the conclusion that this was more likely in higher-claimer companies and for incentives for safety innovations, so validating the conclusion that lower claimers used these practices more than higher claimers. Observations in five out of seven lower claimers, and four out of eight higher claimers supported consistency of the findings about GPS. Managers were less consistent on whether they formally investigated unsafe behaviour, but as the inconsistent responses largely supported the Study 2 conclusion that lower claimers were more likely to use this practice, it was also judged as validated by the Study 3 interviews. Three of the seven lower claimers and only one of the eight higher claimers showed disciplinary procedure documents. Managers were not asked in the interviews about the remaining monitoring or discipline and investigation practices for hours of service breaches. However, one higher claimer and two lower claimers showed procedural documents indicating that investigations of working-hours breaches do occur.

The finding that lower claimers were more likely to provide incentives for safety than were higher claimers was validated by the study. The managers provided consistent answers on this question. Moreover, two lower claimers and no higher claimers provided documentary evidence of their incentive programs.

5.3.6.2 Driver survey findings about monitoring, discipline and incentives Table 5.19 shows that driver interview responses indicate that the trucks in higher-claiming companies were more likely than lower-claiming companies to have GPS monitoring devices fitted to their trucks on the basis that nature of the inconsistent responses by drivers compared to their managers for both lower and higher claimers supported this conclusion. Similar conclusions can be drawn on the use of fuel and speed monitoring devices, and the use of incentives for safety innovations since all of the inconsistent

138

responses between driver and manager supported the conclusions drawn in Study 2: lower claimers were less likely to use fuel and speed monitoring but were more likely to use incentives for safety innovations.

Although drivers in only half of the lower-claiming companies were consistent with the survey responses regarding formal discipline for unsafe behaviour, the nature of the inconsistencies supported the survey finding that lower-claiming companies are more likely to have these disciplinary procedures. Most drivers from higher-claiming companies made consistent responses to their manager’s reports in Study 2, although drivers in two companies reported these procedures contrary to their managers’ responses in the original survey.

On the question about dealing with breaches of working hour limits in the Study 2 survey, if managers in the original survey did not tick any boxes, this was treated as “no” to investigating breaches and disciplining breaches. It was noted that in three of the four lower-claiming companies the managers explained that either working hours limits did not apply to their line of business or that breaches never occur. Drivers in two of these companies voiced agreement to those responses. A driver in another lower- claiming company said, “they would never do this [let drivers work beyond their hours limits].” None of the four higher claimers, neither managers, nor drivers, commented that breaches of working hours did not happen in their companies. While the consistency in Study 2 and 3 findings is reasonably strong, it would be misleading to conclude that the Study 2 finding that higher claimers are more likely than lower claimers to investigate and discipline breaches of working hours, given that many of the lower claimers provided evidence that they prevent breaches rather than react to them.

5.3.6.3 Summary of Study 3 results about monitoring, discipline and incentives In summary, the findings that lower claimers were more likely to conduct pre-start checks, document fatigue management, use experienced driver to check and coach other drivers, formally investigate unsafe behaviour and provide incentives for safety innovations were supported by manager and driver responses in this Study. The Study also confirmed, on the basis of manager and driver responses, that higher claimers were more likely to use GPS tracking, fuel and speed monitoring devices. The conclusions about how companies deal with breaches of working hour limits could not be made.

139 Table 5.17 Consistency for monitoring, discipline and incentives characteristics - numbers of companies showing consistency with the responses by managers in the original survey (numerator) expressed as a fraction of the total number of companies in each sub-group of the survey (denominator) for managers, drivers and documents/observational evidence from the in-depth study. Bolded fractions denote half or less were consistent. In-depth (Study 3)

Large lower-claiming a Large higher- claiming a Small lower claiming a Small higher claiming a (n =3) (n=4)c (n=4)b (n=4)

.J!)C s c s c -~ s -~ ~ ·en ~ c.2 -~ ~ ~ c.2 -~ ~ ·en ~ c .Q c Characteristic Q) Q) - Q) -!!? c .... c ~-ro c !!? c ~-ro c .... c ~-ro Q) ~ ~ ~Q) ~ Q) ::I ~Q) ~ Q) ::I :ti'Q) ~ Q) ::I ~c ~ c ~ c:::-2 ·c::: :2 2: c:::-2 ·c::: :2 2: c:::-2 ·c::: :2 2: c Q) ·c::: Q) ::I ~a; oa; 8.8 ~a; o a; 8.8 ~a; oa; 8.8 032 8 Oo Oo Oo ~~Q) Q)> 0 Conduct pre-start checks •• 3/3 2/3 4/4 1/4 4/4 2/4 4/4 Use GPS tracking 3/3 3/3 1/3 3/4 3/4 2/4 4/4 2/3 4/4 3/4 3/4 2/4 Use fuel monitoring 3/3 2/4 2/3 2/4 Use speed monitoring 3/3 2/4 1/3 2/3 3/4 Use experienced drivers to check/ 3/3 4/4 4/4 4/4 coach drivers Discipline for hours breaches • 3/3 2/4 1/3 2/4 Investigate hours breaches • 2/3 1/3 3/4 1/4 1/3 2/4 1/4 Document fatigue management •• 3/3 2/3 4/4 2/4 4/4 2/4 4/4 1/4 Formally investigate unsafe 2/3 2/3 2/3 4/4 3/4 1/4 2/4 1/3 1/4 3/4 3/4 behaviour Incentives for safety innovations 3/3 3/3 1/3 3/4 3/4 4/4 3/3 1/4 3/4 2/4

•• These characteristics were new discoveries in the survey and were more prevalently observed in lower-claiming companies in site visits.

140 Table 5.18 Manager-provided evidence regarding driver monitoring, discipline and incentives- percentages of companies showing consistency with their responses in the original Survey through managers' interview responses and observations/documentation.

Characteristic Finding from Study % Manager Nature of inconsistencies % Manager Nature of inconsistencies Conclusion 2 survey consistent consistent Conduct pre Lower were more 100.0 100.00 Validated start checks likely to do than Higher Use GPS Lower were less likely 100.0 75.0 Two managers of Higher-claiming companies Validated tracking to do than Higher said that they did not use GPS whereas they said they did in the survey. On-site observations further confirmed the greater use of GPS in H Use fuel Lower were less likely Managers were not asked No mon1tonn to do than Higher conclusion Use speed Lower were less likely Managers were not asked Managers were not asked. No mon1tonng to do than Higher conclusion claimers Use Lower were more 100.0 100.0 Validated expenenced to likely to do than check/tram Higher dnvers Discipline for Managers were not asked. Managers were not asked. No hours breaches conclusion Investigate Managers were not asked. Managers were not asked. No hours breaches conclusion Document 100.0 100.0 Validated fatigue

Formally 71.4 Managers in three Lower-claiming 87.5 In one Higher-claiming company the manager Validated investigate companies described formal investigation had said in the survey that they did this but the unsafe processes applied to safety policy Study 3 interview did not confirm this. behaviour breaches but had not reported formal in the 100.0 75.0 Two Higher claimers had reported to do Validated the but indicated otherwise in

141 Table 5.19 Driver-provided evidence regarding driver monitoring, discipline and incentives - percentages of companies showing consistency with the responses by their managers in the original Survey through drivers' interview responses.

Characteristic % Drivers Nature of inconsistencies %Drivers Nature of inconsistencies Conclusion consistent consistent Conduct pre-start Drivers were not asked. Drivers were not asked. No checks conclusion Use GPS tracking 83.3 Drivers in one Lower-claiming 75.0 Drivers in two of Higher-claiming Consistent company advised their trucks had this companies contradicted survey findings device contra to the their truck did not have GPS. Use fuel monitoring 83.3 Drivers in one Lower-claiming 50.0 Drivers in fou r Higher claimers said these Consistent company advised their trucks had this were on their trucks contrary to the survey. device contra to the Use speed 83.3 Drivers in one Lower-claiming 62.5 Drivers in three Higher claimers reporting Consistent monitoring company advised their trucks had this having these devices contrary to the device contra to the Use experienced Drivers were not asked. No drivers check/coach conclusion Discipline for hours 66.7 Drivers in two Lower claimers said 50.0 Drivers in two Higher claimers said this is No breaches discipline for hours was applied applied contrary to the managers' answer, conclusion contrary to the survey. and two said is not applied contrary to the

Investigate hours 50.0 In three Lower-claiming companies 62.5 No breaches drivers said breaches are conclusion investigated contrary to the survey.

Drivers were not asked. No conclusion Formally 50.0 Drivers in half of the Lower-claiming 75.0 Drivers in two Higher-claiming companies Consistent investigate unsafe companies advised this is done advised this is done contrary to survey behaviour to su results. results. Incentives for 100.0 62.5 Drivers in three Higher-claiming companies Consistent safety said that no incentives are provided the

142

5.4 Differences in consistencies between lower and higher claimers

As reported in Chapter 4, Study 2 survey results showed a total of 37 characteristics found to distinguish between lower and higher truck insurance claimers (see Table 5.20). Seventeen of these characteristics were expected, based on Study 1 findings from prior safety management research. That is, good safety practices were more likely to be found in lower claimers than higher claimers. Twenty of the distinguishing characteristics were not expected to be found – where higher claimers were found to be more likely to have the characteristics than were lower claimers, unlike findings in previous research.

Overall, as illustrated in Tables 5.2 - 5.19, there was more evidence of consistency with survey results found in lower-claiming companies than in higher-claiming companies. The mean percentage consistency for managers from lower-claiming companies was 93% (22.5 out of 24 items tested) compared to 74% (17.7 out of 24 items tested) for managers from higher-claiming companies. Drivers’ reports were less consistent with their managers’ reports of characteristics and practices in their companies, but again, drivers from lower-claiming companies had higher consistency than for drivers from higher-claiming companies (mean percentage consistency 79% for lower claimers and 61% for higher claimers). Furthermore, 16 of the expected 17 characteristics (94%), where lower claimers were more likely to report better safety practices, were validated by Study 3 in-depth investigation findings, contrasting with only 11 of the 20 unexpected findings (55%) were validated. Four unexpected characteristics were not validated by the study. A further five unexpected characteristics were inconclusive due a lack of clear evidence to conclusively confirm or refute them. Lower-claiming companies also exhibited more observational evidence to support the expected findings – visual evidence and supportive documents provided in the company visits. Moreover, the one expected characteristic that could not be confirmed was about location of scheduling and rostering where drivers’ answers suggested that they were not certain about how to describe the location of scheduling and rostering.

In summary, all but one expected safety management characteristic – where lower claimers were more likely to have than higher claimers – were validated. Lower claiming managers were considerably more consistent in their provision of evidence to support what they said in the Study 2 survey than were managers in higher-claiming companies. Just over half of the unexpected findings – where higher claimers were more likely than lower claimers to have a safety management characteristic – were validated by Study 3. Moreover, drivers in lower-claiming companies were consistent with their managers’ survey responses than were drivers in higher-claiming companies.

The next section re-examines the conclusions from the survey in Study 2 against the validation results based on the in-depth investigation study.

143

Table 5.20 Summary of 37 safety management practices confirmed (Y) or not (N) or not tested (-) by the in-depth investigation study compared with Study 2 findings

Recruitment/ Check references, license points, N employment use selection tests

Pay/conditions to pay subcontractors a

Policies y

Accreditations and KPis y

Training y

Communication/ driver input

Monitoring

Discipline/ incentives

5.5 Additional interview questions

In addition to examining the characteristics that were reported in the survey (Chapter 4), the in-depth interviews with managers and drivers included a question about how comprehensive the safety management system is for the company. Managers in the lower-claiming companies more often stressed the importance of managers actively taking responsibilities for managing safety, and they more often demonstrated an interest in continually improving their safety management practices. This was in stark contrast to many managers in higher-claiming companies, who demonstrated that they are not proactive in

144

safety management. The higher-claiming company managers said things to the effect of, “it’s up to the drivers to manage their safety”, “we just try to comply with the law”, “we can’t control that [when asked about drivers’ infringements]” and “we have not had too many major crashes.”

While Study 3 did not specifically focus on safety culture, the in-depth investigations identified some characteristics that distinguished between lower- and higher-claiming companies, which are consistent with Study 1 scientific literature review findings on cultural determinants of safety. Apart from validating specific safety management features found in Study 2, Study 3 provided some evidence that managers and drivers in lower-claiming companies consistently displayed a strong commitment to safety as a priority, with managers demonstrating leadership and responsibility for safety management.

Managers in lower-claiming companies being more proactive in taking responsibility for managing safety than were managers in higher-claiming companies was demonstrated in a number of ways. For example, rather than relying on drivers doing site inspections, a manager in a lower-claiming company advised that a manager conducts a job safety and environment assessment for all new delivery sites. Drivers in another lower-claiming company advised that their manager to looked after their safety of drivers as if they were family. By contrast, managers in higher-claiming companies often complained that they had to face unfair challenges imposed by government or otherwise placed responsibility on drivers or others (customers, depot managers, etc.). For example, one manager in a higher-claiming company complained that a customer did not look after his drivers’ safety, whereas a manager of a lower-claiming company advised that he dropped a customer who would not make the delivery site safer for his drivers. To the question about how WHS data are used, a manager in a higher-claiming company said, “if you want to ask about OHS matters, you should talk with the OHS manager.”

Managers in higher-claiming companies often indicated that there was not much they could do about safe driving and it was up to the drivers and/or the government authorities to take responsibility for this. Moreover, managers and drivers in higher-claiming companies seemed to treat safety management as a task compared with lower-claiming companies where safety was seen as a fundamental quality of the company. A driver in one higher-claiming company (that had been seeking to improve its safety culture) said, “In the past, the former (WHS) person [in place at the time of the survey] fobbed things off” and said that the current management is good.

Most of the higher-claiming companies did not demonstrate a full commitment by managers to lead and actively manage safety. One manager in a small higher-claiming company seemed bemused by the Study 3 interview, saying, “I don’t ever think about safety – until you came here. We have only had four major incidents since 1976 – and two were subcontractors not in our vehicles.” A manager in a large higher- claiming company similarly indicated that they do not worry about safety unless there are costly incidents. Some managers in higher-claiming companies showed a lack of interest in safety. A driver in one large higher-claiming company said, “managers don’t get out on the coalface to see what’s happening.” He said

145 that the response from (older) drivers to safety suggestions by the (WHS) reps is “we’ve always done things this way” and “the young people just thumb their noses.” By contrast, managers in lower-claiming companies consistently demonstrated leadership in conveying the priority they placed on safety, as evidenced by drivers who commented, “If you feel unsafe you just ring up and say ‘I’m not doing it’ – and they work with you to make it safe.”

Managers in lower-claiming companies indicated a greater interest in improving safety management. As an illustration of this, when managers were asked about their response to safety incidents, the higher- claiming companies tended to focus on cleaning up and minimising costs in contrast to lower claimers that wanted to learn from the incident investigation and follow-up with actions to reduce future incidents. By contrast, higher claimers, with the exception of two companies that were undertaking a change process to improve safety management, did not indicate an active interest in improving safety management. For example, one manager replied to the question about how comprehensive his company’s safety management system was, by saying that he tries “to comply with legislation but the paperwork is very difficult.” This contrasts with the comment by a manager in a lower-claiming company who said that they “always strive to improve”.

On the basis of these differences between lower- and higher-claiming companies, it was concluded that, managers taking responsibility and showing leadership in making safety a clear priority, was a characteristic that distinguished lower claimers from higher claimers.

5.6 Discussion

This Study has validated a total of twenty-seven characteristics that distinguish lower-claiming (safer) company safety management features from higher-claiming (less safe) company features. Sixteen of these characteristics were found to be consistent with prior safety management research findings that indicated they would more likely be found in safer versus less safe companies. Eleven of the characteristics validated in Study 3 were not expected based on prior research. For these characteristics, more testing is required in order to reconcile these counterintuitive results. Additionally, a new characteristic emerged from the analysis of the in-depth investigations, consistent with the scientific literature. The lower claimers demonstrated a strong commitment and leadership in making safety a clear priority, whereas higher claimers did not.

The analysis of these consistencies, together with examination of the nature of the consistent and inconsistent answers, provided a sound basis for identifying important safety management characteristics for heavy vehicle road transport companies. The findings on characteristics that were validated by consistent or inconsistent findings are discussed in this Section as well as some additional insights gained in this Study.

146

5.6.1 Validated safety management characteristics (27) The evidence for validating the twenty-seven distinguishing characteristics across all safety management topics is discussed here.

5.6.1.1 Validated truck fleet management characteristics (3) The findings relating to the management of trucks and fleets in the Study 2 survey findings were largely confirmed. The truck management characteristics that distinguished lower claimers from higher claimers were a greater apparent understanding and proactive management of freight carried, specific vehicle safety equipment required, and level and type of vehicle checks and maintenance of trucks, including pre- start checks, needed to assure optimal safety on every delivery journey.

The survey indicated that fleet size was a characteristic difference between higher- and lower-claiming companies, with higher claimers having on average much larger fleets. This was evident also among companies participating in the in-depth investigations. The finding about fleet size may partially explain inconsistencies in higher claimer managers’ and drivers’ responses about their fleets. It may be that communication between sections of the workforce is harder to achieve in larger companies.

Fewer defect notices and days off road due to mechanical breakdowns were also reported in lower- claiming companies, than in higher-claiming companies. This was consistent with the original survey. The confirmation of the finding that lower claimers had fewer defect notices than higher claimers suggests that maintenance practices were superior in those companies. In Study 3 one quarter of managers in higher-claiming companies reported less vigilant truck maintenance practices than they reported in the survey. This may help to explain higher numbers of defect notices accrued in higher-claiming companies compared with lower-claiming companies. Also, through interview and documents, managers in lower- claiming companies demonstrated pro-active maintenance practices, including confirming the survey finding that they were more likely than higher claimers to perform regular pre-start inspection checks. By contrast, managers in higher-claiming companies were more reactive to drivers’ truck repair requests.

The original survey indicated that lower claimers were more likely than higher claimers to consider at least one safety feature in the purchase of trucks. This finding was supported by the in-depth investigation. Of the eight higher-claiming companies, managers interviewed in five of them gave inconsistent answers about considering safety features in the purchase of trucks compared with their survey answers, saying that safety features were not a priority in truck purchasing, or that they only considered features required by law or a customer, or they considered features only that protected the truck (e.g. bull bars) or the freight (e.g. longer tool trays). By contrast all lower claimers consistently indicated in Studies 2 and 3 that they were proactive in ensuring the relevant safety features were included or fitted to trucks, with some lower-claiming company managers also providing documented evidence.

147

5.6.1.2 Validated risk assessment characteristics (4) Managers of lower-claiming companies more consistently than higher claimers reported that they check traffic conditions and other journey-related risks prior to drivers setting out on trips, and this was consistent with the original survey. Notably, one manager of a higher-claiming company indicated that he doesn’t check because he said there is nothing he could do about poor road conditions. Consistent with the survey results, two of the lower claimers checked road conditions and limited speeds on poorer quality roads; whereas none of the higher claimers said that they do this. Another higher claimer advised that he only regularly checks for (load) weight restrictions for regulatory compliance purposes, and mentioned nothing about injury risks to drivers on journeys. By contrast, with regard to journey risk assessments, a company manager in a lower-claiming company said that when drivers must travel to locations where communication access is limited, two drivers will be sent out together, or if communications are possible, there is a ‘call-in, call-out’ system where drivers must call in after one hour.

The Study 2 finding that lower claimers were more likely than higher claimers to carry out site risk assessments was validated in Study 3. The manager in one lower-claiming company who had reported in the survey that site risk assessments were not done, advised in interview that they are done, resulting in a finding that all lower companies investigated in Study 3 do these assessments, whereas only half of the higher claimers reported doing these assessments.

Based on driver responses, there was evidence to support the conclusion that lower claimers were more likely than higher claimers to set time limits on responses to drivers’ safety concerns. Drivers should be aware of the timeliness of responses to their safety concerns. The reports from drivers in this study indicate that even though they differed from their manager’s responses, drivers from lower-claiming companies had little or no problems with managers delaying action on safety concerns, unlike those from higher-claiming companies.

5.6.1.3 Validated recruitment and employment characteristics (4) All lower claimer managers were consistent in reporting that driver recruitment involves checking accident histories of potential recruit drivers. Higher claimers were quite inconsistent in their interview responses on checking accident histories with 62.5% of managers, who had claimed in the survey that they did these checks, advised in interview that they did not do these checks. The finding that higher claimers were more likely than lower claimers to employ drivers over the age of 60 was validated. Interestingly, in interviews, managers in lower-claiming companies advised that they did not simply rule out drivers who were over the age of 60, but instead assessed drivers for fitness to carry out all tasks associated with the job regardless of age.

Study 3 confirmed that lower claimers are more likely than higher claimers to pay drivers by hours worked, as well as to pay them for all hours spent waiting for their trucks to be loaded or unloaded. Apart from one higher-claiming company manager who changed his report from trip/load pay to hourly,

148

and one who changed from hourly pay to trip/load pay, all managers interviewed were consistent with their survey responses. This conclusion was also supported by driver responses.

5.6.1.4 Validated safety policies and accreditation characteristics (5) The in-depth study included a question about fatigue risk policies for both managers and drivers. There was consistency in responses across all sub-groups except for the small higher-claiming companies. There was sufficient consistency between survey and Study 3 findings to validate higher-claiming companies being more likely to have these policies. Further, driver interview responses confirmed the survey findings that higher claimers were more likely than lower claimers to have policies on fleet management and work monitoring.

With regard to accreditations, the consistency in manager reporting and confirmed by drivers supported the survey finding that higher claimers were more likely to be accredited under National Heavy Vehicle Accreditation Scheme (NHVAS) Basic Fatigue Management (BFM) and NHVAS Mass Management. The findings that higher claimers tend to have policies on fatigue risk management and are more likely to be accredited to NHVAS (BFM) are likely to be linked because accreditation requires companies to show that they have fatigue risk and other relevant policies. The fact that accreditation to these Schemes was more prevalent in higher-claiming companies may mean that these accreditations are not necessarily related to safety. This is consistent with the findings of a review of NHVAS conducted by the National Transport Commission, (NTC, 2009, Summary page) that concluded “it cannot unequivocally be said that the policy has improved safety”.

5.6.1.5 Validated driver input into WHS and safety training characteristics (3) The finding that managers from lower claimers were more likely than those from higher claimers to encourage driver input into WHS decision-making was confirmed with strong consistency in managers’ responses, especially from lower claimers. The nature of the inconsistencies shown by managers in higher-claiming companies strengthened the original result. Driver responses also supported this conclusion. Evidence from the in-depth investigations suggested that the communications between managers and drivers in lower-claiming companies were more consistent, effective and safety focused.

With regard to fatigue risk management training, both managers’ and drivers’ interviews were consistent with the Study 2 survey, so confirming that higher claimers are more likely than lower claimers to provide this training. Given that higher claimers were more likely to be accredited to NHVAS (BFM) that requires fatigue risk training of drivers, this could at least partially explain this result. There was a low level of consistency between driver responses and the original managers’ survey on eco-driver training but the nature of the inconsistencies supported the finding that higher claimers were more likely than lower claimers to provide this training.

149

5.6.1.5 Validated monitoring, discipline and incentives characteristics (8) The original survey found that pre-start inspection checks were more often done in lower-claiming companies relative to higher-claiming companies, and was confirmed in the in-depth study. Managers’ vigilance over pre-trip inspections in lower-claiming companies was observed to be greater in lower- claiming companies, with one manager saying, “I would never let a truck out of the yard unless it was in top shape.”

Managers were asked in the in-depth study about how fatigue risks were assessed. All lower claimers showed evidence of having regular and vigilant approaches to assessing, managing and documenting specific fatigue risks associated with the types of work carried out by their drivers. By contrast, managers in higher-claiming companies indicated in the in-depth study that fatigue risk is more of a driver’s than a manager’s responsibility. Thus, the Study 2 survey finding on documenting fatigue management was confirmed. This finding is despite the finding that higher claimers were more likely than lower claimers to have formal policies on fatigue risk management.

Manager and driver interviews as well as on-site observations validate the finding that higher claimers were more likely to use GPS tracking devices compared with lower claimers. Evidence from drivers also provided support for greater use of fuel monitoring devices and speed monitoring devices by higher claimers compared with lower claimers. The in-depth interviews with managers and discussions with drivers suggested that that the greater use of these tracking tools may not be primarily related to safety monitoring. A manager in one higher claimer demonstrated how he uses GPS tracking to check whether a delivery would be on time. A driver in another higher-claiming company said that the speed monitoring by managers was “just another way to get at drivers.”

The finding that lower claimers are more likely, than higher claimers, to utilise experienced drivers to check and coach new drivers, and/or provide remedial coaching and training for drivers exhibiting unsafe practices, was confirmed, with 100% consistency in all managers’ interview responses.

With regard to discipline for unsafe behaviour, the survey indicated that lower claimers were more likely to formally investigate. There was also observable evidence that all of the small lower claimers and two out of three of the large lower claimers had transparent and formal approaches to discipline, for example, these companies had in place a practice of issuing three letters on small breaches before dismissal, and instant dismissal for any major breach or something similar. The third large lower claimer had a less formal approach but the manager indicated that unsafe practices would not be tolerated, saying that drivers who behaved unsafely would be shown the gate21.

There was consistent evidence from managers and supported by drivers that lower claimers used incentives to encourage safety more than higher claimers. Evidence from higher-claiming company managers suggested that they had a different approach to encouraging safety. For example, one higher

21 This means that he would dismiss the driver.

150

claimer manager had reported in the survey that BBQs were offered as safety incentives to drivers, but in the Study 3 interview he said that the BBQs were not incentives but rather an opportunity for drivers to “air their grievances”. By contrast, for example, one lower claimer gave cash rewards for completing job safety analyses, and another removed a safety bonus if a driver breached a safety rule or had an at-fault safety incident. Higher claimers reported providing meals or other non-monetary extras for drivers and the link to safety criteria was unclear. In fact while the survey found that two higher claimers provided incentives for safety innovations, managers and drivers in these companies indicated that the incentives were more related to work criteria than safety related criteria, and therefore discredited their earlier claims that incentives for safety innovations were provided.

5.6.2 Safety management characteristics not validated in Study 3 (4) There were four characteristics that could not be confirmed by evidence collected in the in-depth study. These are described below.

5.6.2.1 Not validated driver recruitment characteristic (1) Managers in higher-claiming companies were found in the original survey to be more likely to check references, license points and test potential driver recruits but in Study 3, the opposite was found. Managers in all but one higher-claiming company said when interviewed that they do not conduct these checks when recruiting drivers, despite their Study 2 survey response that claimed they did do them. Only one Lower-claiming manager contradicted his survey answer saying he did these checks, despite not reporting these checks in the survey. The clear lack of consistency in responding by higher-claiming company managers called into question the validity of the original conclusion.

5.6.2.2 Not validated safety management KPIs characteristic (1) The survey finding that higher-claiming companies had more safety related key performance indicators were not validated by this study. Half of the managers in higher-claiming companies, who reported in the survey that they had safety KPIs, said in the in-depth study that they did not have these in place. Similarly, four of the seven lower claimers (57.1%) who did not indicate in the survey that they had safety KPIs, reported in the in-depth study formal methods of assessing safety performance, including KPIs. The degree of inconsistency and the nature of the changed responses significantly weakened the evidence from Study 2 on this characteristic.

5.6.2.3 Not validated safety training characteristics (2) The original survey showed that on-road driver training and classroom driver training was more common in higher-claiming companies than in lower-claiming companies. This was judged to be invalid from the in-depth study as although no data was available from manager interviews, almost all drivers from higher-claiming companies contradicted their manager’s claims that this type of training was done. The clear lack of support from drivers who should be aware of the types of training offered was sufficient evidence to judge the original conclusion to be invalid. 151

5.6.3 Inconclusive safety management characteristics (6) There were six characteristics found to differentiate between lower and higher claimers in the Study 2 survey that were not sufficiently tested in Study 3 to make conclusions about their validity. These included the characteristics described below.

5.6.3.1 Inconclusive findings on truck fleet characteristic (1) The survey had indicated that higher-claiming companies were more likely to include electronic stability programs (ESP) and/or front or rear underrun protection than lower-claiming companies. While managers were not specifically asked about these features, driver interview data on features on the trucks that they mainly drove did not confirm this finding. However, the validity of this characteristic could not be conclusively determined.

5.6.3.2 Inconclusive findings on scheduling characteristic (1) The Study 2 finding that lower claimers were more likely than higher claimers to centrally schedule and roster drivers was invalidated by Study 3 findings. Drivers in one lower-claiming company advised their company did local scheduling contrary to the survey. Drivers in half of the higher-claiming companies said that their companies scheduled and rostered centrally contrary to their managers’ survey response, and another company’s drivers said scheduling was done from a local base contradicting their manager. Observations in the site visits suggested that the scheduling/rostering location options in the driver survey were not consistently interpreted by drivers – even by drivers working for the same company. So, this characteristic could not be confirmed or refuted.

5.6.3.3 Inconclusive findings on driver pay characteristic (1) The finding that higher claimers were more likely to pay subcontractors by hours worked could not be validated with Study 3 as there were insufficient numbers of companies employing contract drivers in the sample to test this characteristic. Only five companies in the sample employed subcontract drivers, two lower claimers and three higher claimers. They all advised that they paid subcontractors hourly rates.

5.6.3.4 Inconclusive findings on safety training characteristic (1) The characteristic pre-trip inspection training was not tested.

5.3.6.5 Inconclusive findings on driver discipline characteristics (2) The Study 2 findings on investigation and discipline for breaches of working hour limits were not conclusively confirmed because the interpretation of the data found that the reason for many of the lower claimers for not investigating and disciplining drivers for breaches was that the breaches were prevented so that these procedures are unnecessary. Also, half of the lower- and half of the higher-claiming companies did not indicate whether anything was done about these breaches. Therefore, no clear conclusion could be reached.

152

5.6.4 Safety cultural characteristics As discussed in Chapter 3, safety culture is a prominent safety management feature that has been interpreted, from the results of empirical studies, to influence safety outcomes. However, as safety culture is an emergent feature of management systems, it cannot be directly tested by only asking managers in a self-report survey about safety culture in their organisation. The Study 2 survey therefore focussed on characteristics and practices of safety management systems in trucking companies that could be observed. The in-depth study provided an opportunity to obtain more comprehensive information about differences between lower- and higher-claiming companies as it included interviews that allowed more open questions of managers, alternate insights into safety management practices from drivers and documentary and other observational evidence. This assessment found that lower claimers were different from higher claimers on a range of characteristics relating to the culture of the company. In particular, the in-depth study found that managers in lower claiming companies were proactive in taking responsibility for safety management; showed leadership in making safety a priority in the company; and indicated an interest in continually improving safety. By contrast, managers in higher-claiming companies more often relegated safety management responsibilities to drivers, other staff, customers and/or government agencies, failed to demonstrate a commitment to safety management, and showed little interest in improving safety.

These findings are important as they are consistent with the large volume of previous research reviewed in Chapter 3 that highlight the importance of safety culture in safety management. The results of this study add strength to these previous findings as they demonstrate that companies distinguished in terms of safety outcomes differ on a range of safety practices and in how their managers treat safety. Better performing companies have managers who take responsibility and leadership for safety. Of course it is not possible to determine from these studies in detail how these cultural characteristics relate to the observed safety practices, and whether practices occur because of culture or vice versa. Nevertheless, if safety culture plays a role in determining safety outcomes, lower-claiming companies should demonstrate better cultural characteristics and this is what was found.

Further demonstration of the cultural differences between lower- and higher-claiming companies is the difference in consistency of responding to questions on safety practices. An important finding that emerged from this in-depth study is that managers and drivers in lower-claiming companies provided far more consistent responses with the managers’ earlier Study 2 survey responses, than were provided by managers and drivers in higher-claiming companies. An analysis of comparative consistencies, for each characteristic tested (23), shows that, overall the average level of consistency for lower-claiming company manager responses was 94%. Higher-claiming companies’ managers had a consistency average of only 74%. Similarly the overall level of consistency for drivers was also higher for those from lower-claiming companies (79%) compared to higher-claiming companies (61%) for the 24 characteristics on which they were tested.

153

These differing rates of consistency shown by lower claimers compared with higher claimers have some important implications for research methods used in safety management research. In lower-claiming companies both managers and drivers were considerably more consistent with the results of the original managers’ survey. This may reflect differential effects of social desirability bias in cross-sectional studies between better and poorer safety performers. It could also suggest that safety management is more top- of-mind for managers and drivers in better safety-performing companies. Whatever the reason for these differences, they suggest that cross-sectional studies that are not validated in some way may produce results that are less robust than previously thought. Importantly, responses by participants on questions that they know or think they have a weakness, are less reliable than responses by participants that know or think they do not have a weakness.

5.6.5 Limitations of the in-depth investigation There are some characteristics of this study that could limit the conclusions. The sample size for Study 3 was too small to apply statistical significance testing, however, the purpose was not only to explore the consistency of the findings of the survey conducted in the earlier study, but rather to find evidence to confirm or or deny the conclusion from the survey as well as more fully explain survey results from a sample of the survey participants. The sample size of fifteen – and as small as three in the subgroupings – is justified on the grounds established by Baker and Edwards (2012) who advise, “a small number of cases, or subjects, may be extremely valuable and represent adequate numbers for a research project” and this is especially true for hard-to-access populations. The in-depth nature of Study 3 provided significantly more depth of understanding than available in studies involving surveys and similar methodology where large samples can be measured.

The small sample size also meant that the study did not necessarily reflect the variability of the size and nature of business in this industry and perhaps did not capture the full range of safety management practices. For example, the nature of the work carried out by drivers in a local government organisation is very different from a large long distance trucking company, as well as the nature of the risks facing drivers. Nevertheless, many aspects of safety management are generic enough to permit comparisons and these were the aspects that were mainly addressed in the study. Randomisation of candidates was used to minimise the effect of any selection bias. Most importantly, for a validation study, it is not necessary to employ a representative sample.

As the participant managers had already been exposed to similar questions in the first survey, they may have been alerted to the expectations of what might be regarded as “good safety management” and behaved accordingly. This problem was minimised by not providing the managers with the answers they gave in the first survey and by triangulating their responses, drivers’ responses and on-site observations when looking for evidence. Furthermore, the time between the initial survey and interview for managers was around 14 months and there were sufficient inconsistencies between the initial survey and interview responses to suggest that managers were not simply reproducing what they said in the first survey.

154

Finally, there were practices that could not be adequately tested in the in-depth study, particularly where it was not possible to obtain responses from managers and where the driver survey was relied upon and drivers could not or did not provide full or accurate responses. These practices should be tested in future research.

5.7 Conclusions of the in-depth investigation (Study 3)

The in-depth investigation of safety management practices of previously surveyed companies elicited unique information for refinement and validation with respect to the self-report questionnaire survey findings. While it was not expected that survey participants were necessarily deliberately misleading in their answers to the initial survey questionnaire, the site visit and interviews enabled the thesis author to gain a more accurate picture of the safety management practices through a validation process. Moreover, gaining insights from making observations and interviewing drivers and others in the organisations helped to assist a fuller understanding about how the safety management practices influenced work practices and working conditions. The method used to validate safety management survey findings represents a novel method of researching effective safety management practices.

This research found clear differences in management practices between lower- and high-claiming companies, in both large and small companies. The main kinds of differences found were not necessarily those that might be expected to distinguish between large and small companies, though the larger companies generally did have more documentation than the small ones and less direct communication between senior managers and drivers. As expected based on previous research, lower claimers were more likely than higher claimers to equip trucks with safety features, check safety histories of prospective driver recruits, pay drivers for all hours worked regardless of tasks, encourage driver input to safety and respond quickly to driver safety concerns, conduct pre-start inspection checks, utilise experienced drivers to coach newer drivers, document fatigue risk management, carry out formal investigations on unsafe behaviours and provide drivers monetary incentives for safety innovations. Unexpected findings were that higher claimers had larger fleets, policies on fatigue, fleet management and work monitoring, were more likely to be accredited with NHVAS BFM and Mass Management, and use in-vehicle monitoring devices. The unexpected findings about policies could be related to requirements of accreditation. For example fatigue and fleet policies are required under NHVAS rules. Moreover, there was some evidence that the use of in-vehicle monitoring was not for safety purposes.

This study also enabled an assessment of some of the contexts of these safety management practices. For example, there was an observable difference in safety cultures between lower- and higher-claiming companies in terms of managers taking and sharing responsibility for safety, demonstrating personal commitment to safety (by managers and drivers), and being pro-active in identifying risks and ways to improve safety.

155

A total of 27 safety management practices or organisational characteristics were validated in Study 3. Sixteen of these practices were expected and eleven were not expected. These practices related to workplace and driver management. Also, the in-depth study discovered a safety culture characteristic that pervaded lower-claiming companies and was all but absent in higher-claiming companies. Managers in lower-claiming companies as distinct from higher-claiming companies showed leadership and responsibility in making safety a clear priority. So, in all, there were 28 characteristics that distinguished between lower and higher claimers.

5.7.1 Implications and next steps The findings and conclusions from Chapter 4 are the basis of the research in this chapter, but this research (Study 3) has helped to refine the Study 2 survey conclusions by showing where there is confidence that the practices found to distinguish lower and higher claimers are valid, and really do differentiate these two groups. While the practices validated in this study do not necessarily represent the ultimate complete set of validated safety management practices, there is now sound basis for developing a safety management intervention that is likely to achieve safety improvements. This intervention can be empirically tested for its efficacy.

156

Chapter 6: Discussion and Conclusions – Evidence-based Safety Management System (SMS)

6.1 Background to this thesis

Heavy vehicle crashes involving serious injuries and deaths are still occurring in large numbers in Australia and in many other countries. These serious injuries and deaths are preventable. However, as already outlined in earlier Chapters, while much is known about the causes of heavy vehicle related deaths and injuries, there is little knowledge about what kinds and combinations of management practices can best reduce the risk of these deaths and injuries.

At a regulatory level, governments have made efforts to reduce risk factors such as driver fatigue and unsafe driving speeds through legislation and enforcement of driver working hours and permissible speed limits. Moreover, there is a growing recognition of influences on truck drivers from within their companies and the industry more generally. The ‘chain of responsibility’ principle has now been adopted in Australian transport regulations, in addition to occupational health and safety regulations. Some governments have also encouraged the uptake of ‘alternative compliance’ schemes by allowing accreditation to the Australian Trucking Association’s TruckSafe program and/or adoption of the Australian Logistic Council’s National Logistics Safety Code, to be used as a ‘reasonable steps’ defence in occupational injury prosecution cases. However, these schemes have not yet been independently evaluated.

As discussed in Chapter 2, until this study, much of road safety research had been founded on the epidemiological principles embodied in the two-dimensional Haddon Matrix (see page 31). This method of analysis has led to the development of many effective road safety interventions at a macro-societal level, such as random breath testing, road safety audits, and new car assessment programs. Those researching work related road safety have tried to adapt the Haddon Matrix, particularly in broadening the definition of “environmental factors” to include organisational factors (Murray et al., 2009b; Runyan, 1998). However, it has been argued that this approach will always have limited value as it does not enable a dynamic systems analysis (Mooren et al., 2009a; Salmon and Lenné, 2015). Using a systems approach, road injury risk factors could be understood as management system deficits that require a complex system strengthening response involving policy development and enforcement, work and journey planning, safety risk management education and other systematic management actions. The management system is characterised by a fluid interaction among managers, drivers and others within an organisation, as well as these interactions being influenced by external environmental systems (Stuckey et al., 2007). A paradigm shift in road safety thinking is needed and a systems approach provides that opportunity to think about solutions in a different way from the traditional method of applying a static epidemiological analysis of injury contribution factors.

157

6.2 Aims of this thesis

This thesis set out to develop an evidence-based safety management system suitable for heavy vehicle transport operations, by identifying the practices that distinguish companies that have poorer safety records and those that have better safety records. Insurance claim rates were used as a proxy measure of safety outcomes and to distinguish poorer and better performing companies. Companies with lower insurance claim rates were judged to have better safety records than companies with higher claim rates.

The overall aim of the research was to develop a set of characteristics and practices relating to safety management that, if implemented, would be likely to produce better safety outcomes. Two specific objectives of this research were to: 1. identify the distinguishing characteristics and practices of heavy vehicle transport operating companies with good safety records and those with poorer safety records; and 2. develop an evidence-based safety management system suitable for companies that operate heavy vehicles for transport of goods that will achieve good safety outcomes.

6.3 Approach taken in this thesis research

This thesis has taken a systematic approach to gathering evidence about the practices of truck operators that distinguish companies with good safety results (lower crash outcomes) from those with poorer safety results. Instead of starting with injury factors identified from past crashes, the emphasis of this study was to identify management and organisational practices associated with good and less good safety outcomes. In other words, it examined comparatively approaches to managing injury risk rather than dissecting the risk itself and trying out ways of reducing the risk. The starting point was differing levels of risk environments and the discovery of differing features of risk management systems. The research was designed as three consecutive studies, each building on knowledge gained from the study carried out before.

Study 1 was a systematic scientific literature review that aimed to identify effective safety management practices and organisational characteristics associated with good safety performance in any industry sector. The aim was to find existing safety management practices that have shown evidence-based statistically significant associations with safety related outcomes. The Study produced a list of practices that might be expected to have links to positive safety outcomes (lower crash and injury rates) when applied to the heavy vehicle transport sector. The results from this Study were used to inform the choice of items to be tested in a survey of heavy vehicle transport operating companies (Study 2).

Study 2 was a survey of companies operating heavy trucks to compare practices of those companies that have lower (truck) insurance claim rates with those with higher (truck) insurance claim rates. Insurance claim rates were used as a proxy representation of crash outcomes. The objective of this survey was to

158

identify organisational and management characteristics that differentiated between companies with better and less good safety performance in terms of these measures.

Study 3 was an in-depth qualitative investigation designed to validate the findings of the Study 2 survey. This research attempted to confirm or refute practices found in the previous study to distinguish between companies with lower and higher insurance claim rates identified in the survey. The collection of data involved interviews with managers who participated in the previous survey, collection or sighting of documents relevant to the practices of interest, making visual observations and interviewing drivers to find evidence of those distinguishing practices.

In this Chapter, the analysis of all data collected in Studies 1, 2 and 3 is used to develop a safety management system that is suitable for implementing in companies that operate heavy trucks that should reduce crash outcomes. This research is important because there has been no previous attempt to develop an SMS for this industry. Furthermore, there have been few attempts to develop an evidence- based SMS with a validated set of safety management characteristics for any industry.

6.4 Summary of Study 1, 2 and 3 findings

This Section presents an overview of Study 1, 2 and 3 results.

6.4.1 Strategic literature review (Study 1) findings Knowledge gained from the scientific literature identified a number of specific safety management interventions associated with good safety performance. In order of most to least number of relevant studies found, the safety practices shown to have significant links with safety outcomes included: management commitment/safety climate (30 studies), worker input to WHS, safety communications (21 studies), vehicle/workplace conditions (13 studies), safety training (12 studies), scheduling/journey planning/work pressure (11 studies), safety management systems/accreditation schemes (9 studies), safety policies/procedures/enforcement (8 studies), financial performance/pay systems/pay rates/unionisation (8 studies), risk analysis and corrective actions (8 studies), incentives (7 studies), size of organisation/truck fleet/freight type (6 studies), worker characteristics/driver attitudes/behaviours/health (4), hiring practices/driver retention/return to work policies (4), and prior safety violations, crashes/incidents (2).

This wide range of characteristics was found from studies using limited research methodology. Most of the studies were cross-sectional surveys that did not provide a clear case for the direction of influence of the characteristics studied. It was therefore necessary to design a study to investigate associations between safety outcomes and safety management characteristics.

159

6.4.2 Survey of managers (Study 2) findings The design of Study 2, a survey of managers, was to administer a questionnaire to those in companies with lower and higher insurance claim rates. All of the Study 1 characteristics, except for management commitment/safety climate and financial performance, were included in the survey questionnaire (Appendix B). Management commitment/safety climate could not be accurately tested by asking managers about these characteristics as social desirability in survey responses would render the findings questionable (Grimm, 2010). Management commitment and safety climate are characteristics that are seen when there is a sufficient accumulation of a range of good practices (DeJoy et al., 2010; Williamson et al., 1997). Therefore, the survey, focused on tangible verifiable safety management characteristics, was expected to shed some light on management commitment to safety when investigated in Study 3. Also, as discussed in Chapter 4 (page 87), questions on profitability of companies were excluded due to likely sensitivities about companies revealing their financial position.

This survey identified 37 characteristics found to distinguish between lower and higher truck insurance claiming companies. Seventeen of the characteristics were expected, based on the Study 1 findings from the scientific safety management literature. The findings on 20 of the characteristics were not expected, based on the fact that the findings were not consistent with what good safety might be expected to look like. For example, the Study 2 survey found higher claimers had more policies, did more training of drivers and did more driver-monitoring, all of which have been found to be associated with safety in previous research.

The value of this study was to look at characteristics relating to specific independent variables in good performing and poorer performing companies. Thus it actively compared two clearly defined groups of companies rather than just surveying a poorly described set of companies often with no knowledge of safety outcomes. However, due to the limitations of self-report surveys, as well as a number of unexpected findings, it was important to carry out a further study to validate the survey findings. This is also a unique characteristic of the research design presented in this thesis.

6.4.3 In-depth investigation validation of survey (Study 3) findings Study 3 was designed to investigate the validity of Study 2 survey findings through an in-depth audit of these findings involving interviews of a sample of managers who participated in the original survey, a survey/interview of drivers, on-site observations and documentary review. Figure 6.1 shows a summary of the findings from the Study 3 in-depth investigation of the 37 characteristics that distinguished lower and higher claimers in the Study 2 survey. As shown in the figure, the Study 3 in-depth investigation (as reported in Section 5.6) was able to validate 27 characteristics (73%) of the 37 that were found in Study 2 to differentiate between lower and higher claimers with respect to safety management. These included 16 characteristics that, based on the scientific safety literature, were expected - that is, characteristics that are thought to improve safety management, and were more prevalent in companies with lower insurance claim rates. These 16 characteristics represent specific ways in which lower-claiming companies managed

160

safety risks associated with the work environment, the drivers and communications. These are presented in Table 6.1. The study also validated eleven distinguishing characteristics that were unexpected or inconsistent with what might be expected based on previous research on safety management, including higher claimers having more policies, accreditations, doing more training, and monitoring than lower claimers. These were policies and practices that have been thought to represent good safety management, and yet higher claimers were found in this study to be more likely to have them in place. These results call into question the value of these policies and practices for safety management.

Four of the 37 characteristics were found in the investigation not to be validated. All of these were characteristics not expected of poorer safety performers on the basis of prior research. In the case of driver recruitment checks and safety KPIs, managers in higher-claiming companies admitted that they do not have these practices in place. With regard to safety training, the drivers in higher-claiming companies reported that these programs were not offered by their companies. Therefore, it was concluded that higher claimers were not more likely than lower claimers to have them. These findings mean that these characteristics cannot be included in the safety management system presented in this thesis.

For six characteristics there were insufficient data from the in-depth study to make a conclusion about their validity, and therefore they were deemed to be inconclusive. Of the six characteristics neither validated, nor invalidated, five were findings from Study 2 that were not expected based on prior research. The validation of one expected finding - that lower claimers were more likely than higher claimers to schedule and roster drivers from a central base - could neither be confirmed, nor refuted. The only expected finding that could not be validated was central versus local scheduling and rostering. The testing of this variable seemed to be confounded by smaller companies where drivers in the companies with only one site said that this was a central and a local practice. Two of the five unexpected findings concerning actions taken when drivers breached working hours were somewhat confounded by a number of lower-claiming companies that indicated only that the problem does not arise, and hence did not have practices in place to deal with it. The other three unexpected findings were not sufficiently tested in Study 3 to make conclusions about them. The lack of evidence to support these characteristics does not mean that they are refuted. Additional research could be applied to further test these characteristics. However, these characteristics cannot be included in the evidence-based safety management system (SMS).

The in-depth investigation also revealed an additional characteristic relating to the style or culture of the company, which again differentiated lower- from higher-claiming companies. This characteristic was that managers in lower claimers, as distinct from higher claimers, demonstrated acceptance of responsibility, leadership and proactive approaches to safety management. Study 3 found strong evidence that managers in lower-claiming companies, but not higher-claiming companies, were more vigilant and proactive in their efforts to ensure they were doing all that was possible to assure safe transport operations. Higher claimers, by contrast, often made comments to the effect that they place safety

161 management responsibilities solely on the drivers. This additional characteristic was therefore included in the final evidence-based SMS.

...------,. ...------,. Examined New characteristics characteristic (n=37) (n=1)

Validated Not validated ~ Inconclusive characteristics characteristics characteristics ~ (n=27) (n=4) (n=6)

Unexpected Expected Unexpected Expected Unexpected ~ characteristicsExpected characteristics characteristic characteristic characteristic characteristic (n=16) ~ (n=ll) (n=O) (n=4) (n=l) ~ (n=S) Figure 6.1 Summary validation findings from Study 3

At the end of the survey and validation process, a set of 17 characteristics was revealed: 16 validated characteristics and one arising from the validation process itself, all of which had been shown to distinguish lower- from higher-claiming companies, and where good practices were more likely to be found in lower-claiming companies. These characteristics formed the basis of an evidence-based SMS that will be discussed in more detail in the next section.

6.5 Set of evidence-based safety management characteristics

The culmination of the three studies provided a set of characteristics of trucking companies that are at least associated with lower claims for safety related incidents. These characteristics form the basis for the development of an evidence-based safety management system. This set of 17 characteristics are most likely not an exhaustive set of elements or characteristics/practices of an SMS. There may well be others, but this set distinguished lower- and higher- claiming trucking companies. Unlike many other studies of the potential components of safety management systems, the elements of this SMS were validated by triangulating the results of three research studies. This Section provides a summary of evidence-based safety management characteristics that were identified through synthesising the findings of all three Studies.

The characteristics were then reframed as implementable safety management practices and grouped into logical management items. This involved combining two characteristics relating to driver remuneration into one. That is, the characteristics "pay by time worked" and "pay to wait" were combined as "drivers are paid for all hours worked regardless of task or activity". Also, "pre-trip inspection checks" were linked to having "fewer defect notices" for practical purposes, forming a single practice, "maintenance and pre-trip vehicle checks ensure that trucks are in safe conditions for all trips". This resulted in a total of 14 auditable SMS management practices, as shown in Table 6.1. The 14 practices were then grouped into topic areas and the topic areas further grouped under headings:

162 • Risk assessment and management (6 practices) - covering topics relating to fleet, environment and job risk safety management; • Driver risk management (6 practices) - covering driver employment, remuneration, training, monitoring, discipline and incentives; and • Safety culture management (2 practices) - covering communication management.

Table 6.1 Evidence-based safety management characteristics and practices Group ITopic I Study finding - validated I Evidence-based management practices characteristics Fleet Safety features in choosing All appropriate safety equipment, including c vehicles safety features on trucks, is provided -Q) E Q) Fewer defect notices Maintenance and pre-trip vehicle checks C) C'G Pre-trip inspection checks ensure that trucks are in a safe condition c C'G for all trips E Journey risk Check traffic conditions Route risk assessments are done for all ~c C'G assessment Speed limiting on poorer delivery journeys c quality roads -Q) E 1/) Site risk Safety audits at own sites Site and job risk assessments are regularly 1/) Q) carried out 1/) assessment 1/) C'G Monitoring Document fatigue Monitor fatigue management practices ~ 1/) management ii2 Response to Time limits on response to All managers respond quickly to safety safety concerns drivers' safety concerns concerns raised by drivers Recruitment/ Check accident history Recruitment criteria focus on safe driving employment records Fewer drivers over 60 Driver fitness is assessed to ensure drivers' abilities to safely carry out all job c -Q) duties E Q) C) Pay/conditions Pay by time worked (not by Drivers are paid for all hours worked C'G c trip or load) regardless of the task or activity C'G E Pay to wait ~ 1/) ·;:: Training Experienced drivers Training for drivers is based on individual

....Q) check/coach other drivers tuition by experienced safe drivers ·;::> Discipline Formal investigation of Identified unsafe behaviours are formally c unsafe behaviour investigated Incentives Offer incentives for safety Drivers are given incentives, including innovations monetary incentives, clearly linked to work safety efforts Communication Encourage driver input into Managers encourage driver input to WHS ~~ WHS decision-making Q) :::1 -a~ Show management Managers take responsibility and show en~ commitment to safety leadership in making safety a clear priority management These evidence-based practices were used to construct a safety management system. The fourteen management practices are more fu lly described, together with the rationale for including them in the SMS in the next three sections.

163 In providing the rationale for these safety management practices, evidence from the original research conducted in Studies 2 and 3 forms the foundation of the rationale for their inclusion. Supplementary evidence from previous empirical research, documented in the scientific safety literature, is incorporated to provide further context to strengthen the rationale for including each practice and to provide suggested methods for measurement.

6.5.1 Evidence-based safety risk assessment and management practices (6) For the first grouping, risk assessment and management, there are six evidence-based practices. These safety management elements focus on assessing and managing risks associated with freight, trucks, sites, journeys and improving safety of the workplace and work system by analysing and remediating risks. These are described and justified in sub-sections 6.5.1.1 – 6.5.1.6.

6.5.1.1 All appropriate safety equipment, including safety features on trucks, is provided The set of three studies found converging evidence to support the conclusion that a safety management system (SMS) should include the practice, that all appropriate safety equipment, including safety features on trucks, is provided.

A myriad of optional safety features can be purchased with new trucks, including electronic stability control, speed and lane assist devices, underrun protection, integrated seatbelt and suspension seat, anti-lock braking systems, tyre pressure monitoring systems, non-slip steps and GPS. Prior research has found that these features can either assist the driver to avoid crashes/accidents or minimise the harm in the event of a crash/accident, and should be considered for inclusion when truck purchases are made (Langwieder et al., 2001; Mahmood et al., 2006; Muresan, 2007). The Study 2 finding that managers in lower-claiming companies were more likely than those in higher-claiming companies to consider at least one safety feature was validated in Study 3. The current research also provided ideas on how to achieve this safety management practice. Managers in lower-claiming companies explained how they take into consideration the freight carried and the needs of each individual driver when purchasing and equipping each truck in their fleets.

Ensuring that all appropriate safety equipment is provided is a practice that can improve safety and can be readily audited. It is also possible to audit a company’s truck fleet to assess whether or not the company has implemented this item.

6.5.1.2 Maintenance and pre-trip checks ensure that trucks are in a safe condition for all trips The research also found converging evidence to support the conclusion that the practice of maintenance and pre-trip checks ensure that trucks are in a safe condition for all trips, is an important component for inclusion in the SMS.

Prior research has found that good safety management dictates that, before each driving journey, the condition of the truck and its equipment and load be checked for soundness and incident prevention

164

(Cantor et al., 2010; Friswell and Williamson, 2010), and maintenance be regularly attended to. Participating companies in the current research provided suggestions for ensuring vehicle maintenance is a priority. Study 3 found that while managers in both lower and higher-claiming companies appreciated the need to maintain the trucks in good working order, the lower claimers were more inclusive of driver input and gave more individual attention to each truck and driver combination.

Ensuring that trucks are in a safe condition for every delivery journey could involve the practice of managers and drivers doing pre-trip vehicle checks, and keeping maintenance records on each individual truck. Again, it is possible to audit a company’s truck fleet to assess whether or not the company has implemented this item.

6.5.1.3 Route risk assessments are completed when planning all delivery journeys The results of the three studies indicate that route risk assessments are completed when planning all delivery journeys is likely to improve safety. Study 3 confirmed that lower claimers were more likely than higher claimers to check traffic conditions and limit speeds on poorer quality roads. In a study of light/short haul truck drivers, a common concern of these drivers were the potential hazards that they may encounter on the roads they travel (Friswell and Williamson, 2010), suggesting the importance of route risk assessment. A route risk assessment can involve checking road, bridge, traffic and weather conditions as well as other possible hazards such as animals on the road or road works.

Rest areas and parking areas should be identified and plans for stopping to rest should be included in the trip plan (Sabbagh-Ehrlich et al., 2005). Contingency planning may involve identification of alternative routes, and stopping areas, and making provisions for drivers to contact the manager in the event of any hazard or delay encountered. In this research, one lower claimer advised that, whenever a task involved a driver travelling to an area not within satellite phone contact, an additional driver would be sent in the vehicle. There are now quite sophisticated tools for assessing truck route risks (Cassini, 1998; Chen and Chen, 2011). An assessment of the risks that the driver may encounter en route should be made and a trip plan prepared, highlighting the risks that will need to be managed by the driver and the manager. These trip plans can be documented and audited.

6.5.1.4 Site and job risk assessments are regularly carried out Prior research discussed in Chapter 3, together with the original survey and its validation, support the inclusion in an SMS of the practice, site and job risk assessments are regularly carried out. Both the physical and psychosocial work environments can influence safety and injury outcomes (Bjerkan, 2010; Cui et al., 2013; Geldart et al., 2010). It follows that in the road freight transport industry, drivers should be alerted to any potential hazards on any site visited. Job risk assessments should also identify any safety risks associated with particular tasks that may be performed by the driver (Wachter and Yorio, 2014a). Site risk assessments or job safety analysis procedures were more prevalent in companies with lower insurance claims and they provide ideas for encouraging and managing regular risk assessments. One

165 lower claimer provided a monetary incentive for completing risk assessments each week. Another lower claimer showed documents outlining specific risk assessment procedures for every customer site. Some of the managers in lower-claiming companies provided copies of job and site risk assessment forms. These forms and procedures are readily implementable and auditable.

6.5.1.5 Monitor fatigue management practices The combination of findings from prior research and Studies 2 and 3 regarding the links between worker/driver fitness and safety performance provides evidence to support inclusion of monitor fatigue management practices in the safety management system.

Fatigue is a major driving risk for truck drivers. The testing of driver performance fitness on a simulator22 in New Zealand found that 9% of drivers did not meet the pre-defined driver fitness performance criteria (Baas et al., 2000). The implications suggest that driver fitness checks, in some form, are warranted.

Moreover, safe journey plans are required by a determination of the, now disbanded, Australian Road Safety Remuneration Tribunal. Delivery times may be extended due to a variety of factors that may cause travel delays. It is important, whether or not there is a delay or hours extended, that a driver must be urged to stop and rest when tired. While the recording of working and rest hours are required of every driver by law, the manager under Chain of Responsibility provisions is also responsible for ensuring that drivers are fit for duty and do not exceed safe working hours, regardless of regulatory limits on hours of work.

Study 3 found that while lower claimers were vigilant in checking driver fitness prior to delivery journeys, higher claimers were not vigilant. Fatigue and fitness for duty are documented and are always checked by managers in lower-claiming companies, but not always by those in higher-claiming companies. The pre- trip checks include ensuring that both driver and vehicle are fit for the tasks assigned each day. Higher claimers left the responsibility for risk assessment largely to the drivers. For example, a manager in a higher-claiming company said that working out breaks during the journey is up to the driver, whereas a manager in a lower-claiming company advised that he sits down with each driver and together they plan the tasks.

The Study 2 survey did not detect differences between lower and higher claimers in checking driver fitness through medical or other tests. However, in Study 3 it was discovered through managers’ interviews and policy documents that drug and/or alcohol testing is more prevalent in lower-claiming companies (50%) compared with higher-claiming companies (25%). In addition, one lower claimer visually assesses drivers each day and, if drugs or alcohol are suspected, the driver is not permitted to drive that day. Also, another lower claimer advised that he was considering the introduction of alcohol ignition interlocks on trucks. A manager in one of the higher claimers reported that the company conducts tests drivers for cannabis on

22 Criteria validated in California. See Stein, A. C., Parseghian, Z., Allen, R. W., Miller, J. C. 1992). High risk driver project: Theory, development and validation of the Truck Operator Proficiency System TOPS) Vol. 2: Report. Hawthorne, CA: Systems Technology, Inc. Technical Report 2417-1).

166

recruitment but that he personally did not think this was fair, indicating that many drivers did not come back for interview after their tests (as so many drivers are cannabis users). By contrast, two of the lower claimers conduct random tests of drivers.

Signed forms indicating fitness for duty as well as results of random drug and alcohol testing can be kept on record for auditing purposes.

6.5.1.6 All managers respond quickly to safety concerns raised by drivers Study 3 validated the Study 2 finding that managers in lower-claiming companies are more likely than managers in higher-claiming companies to put time limits on their responses to drivers’ safety concerns. Regardless of whether there were formal policies about time limits, drivers in lower-claiming companies consistently said that managers respond quickly to any safety concerns they raise. The practice, all managers respond quickly to safety concerns raised by drivers is important to include in a safety management system. This practice can be documented as a management procedure and audited or checked by periodic staff surveys.

6.5.2 Evidence-based driver risk management practices (6) Under the second grouping, driver risk management, there are six management practices. This topic is about how to manage risks associated with driver behaviour. The practices are described, together with research evidence to support each element and its measurement below.

6.5.2.1 Driver recruitment criteria focus on safe driving records This study formed a convergence of research evidence to confirm the importance of using driver recruitment criteria that focus on safe driving records. From a safety perspective, testing the safety risk background and risk propensity of drivers, including accident histories, has been found to be an important safety management practice (Darby et al., 2009). The recruitment of drivers by lower claimers involved more comprehensive safety focused assessment than did higher claimers’ recruitment processes, including checks on driving accident histories. Higher claimers consistently checked references, but were less likely to check accident histories or safety records of drivers when recruiting. Managers from lower- claiming companies who recruit drivers check safety records and accident histories and this should be implemented. This can be recorded on personal files of drivers and audited.

6.5.2.2 Driver fitness is assessed to ensure drivers’ abilities to safely carry out all job duties The triangulation of prior research and Studies 2 and 3 found that while higher claimers were more likely to employ drivers over the age of 55 than were lower claimers, the important physical aspect of drivers relating to safety was their fitness and abilities to safely carry out all job duties. Assessments were carried out at the recruitment stage and throughout the employment tenure for each driver to ensure that drivers are not at risk of injury through lack of fitness or capacity to carry out work.

167

Regardless of age, driver recruits should be assessed for their ability to carry out the tasks associated with the job in a safe manner (Guest et al., 2014). The in-depth investigation found that managers in lower- claiming companies did not rule out drivers on the basis of age, but rather assessed drivers on the basis of their fitness for tasks required of the drivers. It was noted that some of the lower claimers employed drivers to do more than driving related work. Sometimes the work of the company involved heavy labour requiring workers with sufficient physical strength to carry out these tasks. So, while lower claimers were less likely to employ drivers over the age of 55, the rationale for employment did not exclude candidates or drivers by age, but rather an assessment of fitness for all tasks including tasks other than driving.

Driver fitness assessments can be documented at the hiring stage, and updated throughout the employment tenure. One company, Roche Australia, has implemented an online driver risk assessment program and used data from these assessments to inform policy development and targeted interventions (Murray et al., 2012). The authors concluded that the driver assessment program contributed to a 24% reduction in motor-vehicle insurance claims between 2004-2009. Tools exist to assist companies to assess and monitor driver fitness, knowledge, skills and risk propensities23. These assessments can be ongoing and monitored for use in SMS improvements and auditing.

6.5.2.3 Drivers are paid for all hours worked, regardless of the task or activity Convergence of prior and current study findings confirms the importance of ensuring that drivers are paid for all hours worked, regardless of the task or activity. Drivers not paid for all hours can tend to make up for their loss by working extra hours or extra jobs. Also, importantly, if they are paid on the basis of productivity, they may be more likely to take risks such as speeding and driving long hours (Hensher and Battellino, 1990; Mayhew and Quinlan, 2006). It is argued, in the literature cited, that paying drivers for all hours worked gives them stability and certainty of income and they are less likely to work in an unsafe manner. A number of studies into pay methods have demonstrated that the method of driver pay influences safety outcomes (Hensher and Battellino, 1990; Hensher et al., 1991; Williamson, 2007).

The Study 2 survey finding that lower claimers were more likely than higher claimers to pay drivers for the time worked instead of by the trip or truckload was confirmed in Study 3. The validation Study also confirmed that lower claimers were more likely to pay drivers for the time they spent waiting to be loaded or unloaded.

The practice, to pay drivers for all hours worked regardless of task or activity, should be implemented and monitored as an important measure to improve safety.

6.5.2.4 Training provided for drivers is based on individual tuition by experienced drivers The SMS practice, training provided for drivers is based on individual tuition by experienced drivers, was found in this research to distinguish between lower and higher claimers. In general, the research results on safety benefits from driver training are mixed (American Transport Research Institute,

23 See www.virtualriskmanager.net for information about the tool used in the Roche study.

168

2008) and this research found there was little difference between lower and higher claimers in respect of the amount of safety related training provided. In fact higher claimers were found to be more likely to use standardised driver training courses. However, lower claimers were more likely to provide safety training tailored to address the specific risks of the job tasks performed and to be based on perceived drivers’ skill deficits, through using experienced drivers to train or coach less experienced drivers or drivers found to have skill deficits. This approach is consistent with other research (Robotham, 2001), that identifying specific safety learning needs related to job tasks is an important training success factor. Therefore, training provided for drivers based on individual tuition by experienced drivers should be implemented to improve safety outcomes. The training assessments and specific training provided can be documented and audited.

6.5.2.5 Identified unsafe behaviours are formally investigated The evidence from the Study 2 survey, and validated in Study 3, supports the inclusion of the SMS practice, to formally investigate unsafe behaviours. Probst and Estrada (2010) found that employees’ perceptions of safety policy enforcement is a predictor of accidents and accident reporting. If discipline is not consistently applied to all drivers the actions may be seen as excuses to punish drivers for simply being unpopular with management. In the current research lower-claiming companies had in place more consistent approaches to safety related disciplinary investigations than higher claiming companies. The practice, identified unsafe behaviours are formally investigated, can be codified in company procedures and audited as well as tested through periodic staff surveys.

6.5.2.6 Drivers are given monetary incentives clearly linked to work safety efforts Prior research evidence that incentives for safe behaviour and safety innovations are effective safety management practices, was supported in Study 2 and validated in Study 3. The finding was that lower claimers, and no higher claimers, give drivers monetary incentives clearly linked to work safety efforts. Incentives or bonuses provided for safe driving can be effective in reducing crashes (Banks, 2008; Gregersen et al., 1996) and thus costs to a company. If incentives or extra rewards are provided to drivers, they should not encourage underreporting of incidents. Positive incentives can be used to encourage employee participation in WHS activities or promote safe behaviour. The objective of incentive programs is to convey to workers that their contributions to safety improvement are valued by the organisation (Fernandez-Muniz et al., 2007b).

Where lower claimers provided additional incentives for safety, the financial incentives were clearly linked to safety criteria, for example for completing job risk analyses, or end of year bonuses for safe driving. By contrast, higher claimers reported providing meals or other non-monetary extras for drivers but the link to safety criteria was unclear. In fact, a manager in a higher-claiming company, contrary to his Study 2 survey answer, advised that the BBQs provided for drivers were not meant as a safety incentive, but rather to give them an opportunity to “air their grievances.”

169

Whatever incentive is offered, it should be clearly linked to safety advancement. This practice can also be documented and audited.

6.5.3 Evidence-based safety culture management practices (2) Under the third grouping, safety culture management, there are two distinguishing communication practices identified from the study. This Section is about managing the safety culture of the organisation. The elements of safety communication are described, together with the evidence on which they are based, in the safety management system in subsections below.

6.5.3.1 Managers encourage drivers to have input to WHS decision-making The triangulation of Study 1, 2 and 3 research results strongly support the inclusion of the practice, managers encourage drivers to have input into WHS decision-making, in the SMS for heavy vehicle transport operations. Two longitudinal studies have shown that implementing interventions involving driver discussion groups focusing on safety risks and safety ideas can reduce crashes (Gregersen et al., 1996; Salminen, 2008), suggesting that encouragement of driver input into WHS decision-making can improve safety outcomes. Two other studies provide evidence that active participation by employees in decisions about maintaining or improving safety is associated with lower injury and accident rates (Vredenburgh, 2002; Wachter and Yorio, 2014a). In fact, there is evidence to suggest that the combination of management commitment and worker participation is an important measure of safety climate (Dedobbeleer and Beland, 1991). Having studied safety for remote workers, Huang et al (2013b) argue that it is especially important to have effective channels of communication when truck drivers, and other employees, work largely in isolation from managers and other workers. Given that drivers often spend long periods of time alone, they may feel that they cannot communicate about safety issues that concern them, i.e. that they must assume sole responsibility for their own and others’ safety. Indeed, during the in-depth investigations the drivers in both higher- and lower-claiming companies expressed a need for this communication.

Practices like providing opportunities for driver input to safety decision-making distinguished companies with lower claim rates from companies with higher claim rates. Managers in higher-claiming companies reported that they were more likely to set criteria and rules for vehicles and drivers without consultation with drivers, than did managers in the lower-claiming companies. It was also observed in Study 3 that lower claimers seemed to focus more strongly on proactive risk assessment, ensuring that rules are agreed, and consulting drivers on safety issues. Consultation procedures and clear communication channels across the company can ensure that changes in WHS policies and procedures involve drivers. These practices can also be documented and audited and can also be reviewed together with staff satisfaction surveys.

170

6.5.3.2 Managers take responsibility and show leadership in making safety a clear priority While Study 1 found 30 studies demonstrating the importance of building a strong safety culture, the items in this characteristic, such as management commitment to safety, could not be accurately tested in the Study 2 manager survey. However, the in-depth investigations (Study 3) found clear and consistent differences in that, in lower-claiming companies, managers were found to take responsibility and show leadership in making safety a clear priority, whereas managers in higher-claiming companies did not.

The fact that managers in higher-claiming companies were far less consistent in self-reported safety management practices than were managers in lower-claiming companies, suggests that their knowledge of safety practices in their company was poorer, possibly due to lower commitment to safety management. Moreover, drivers in higher-claiming companies compared with drivers in lower-claiming companies were similarly far less consistent with their managers’ descriptions of their companies’ safety management. This could mean that communications about safety were not effective in these organisations, or that drivers do not regard safety as a priority in their company and therefore do not pay much attention to safety management practices.

It can be concluded, from the Study 3 results, that the heavy vehicle transport companies with lower insurance claim rates tended to take a more active and substantive approach to managing safety in their organisations, whereas the higher-claiming companies were found to take a more passive business-as- usual style of managing safety. Managers in lower-claiming companies were found to more fully accept responsibility for safety management, whereas many managers in higher-claiming companies complained that they had to face unfair challenges imposed by government or otherwise placed responsibility on drivers or others (e.g. customers, depot managers). Managers in five higher-claiming companies used phrases, in relation to safety management tasks, such as: “drivers are supposed to do this but they don’t”; “what can I do?”; “we can’t control that”; and “we just tell them to do everything by the book.” In contrast, managers in lower-claiming companies check traffic and journey conditions and assist journey planning to ensure adequate rest breaks for drivers, e.g. booking their accommodation in advance.

Based on their research on SMS practices in the airline industry, Chen and Chen (2014) argue that when employees perceive that managers exhibit safety leadership they are more likely to be motivated to comply with safety rules. Evidence from the in-depth investigations found that the communications between managers and drivers in lower-claiming companies were more consistent, effective and safety focused. The safety manager and operations manager in one lower-claiming company were clearly aligned in their mutual commitment to safety – nearly finishing each other’s sentences – and driver interviews supported this observation. By contrast, drivers in higher-claiming companies were not satisfied with safety communications by managers, and sometimes said that their managers contradicted safety messages provided by safety representatives or their policies, undermining their safety messages. Drivers and managers alike in higher-claiming companies demonstrated a much lesser interest in safety, with one 171 manager saying he hardly ever thinks about safety, and managers in a two other higher-claiming company saying it would be difficult to get drivers interested in safety.

The in-depth investigation (Study 3) found that lower-claiming company managers demonstrated genuine and consistent leadership and encouragement of safe behaviours; whereas among higher claimers the managers often shrugged off their role in setting an example for desired safety behaviours. One WHS representative in a higher-claiming company advised that a senior manager didn’t wear a safety vest in the depot even after he was reminded of the policy requirement to do so. By contrast a lower-claiming company manager impressed drivers, not only with strong safety leadership as a manager, but was also active in a local road safety advocacy group. A manager in another lower-claiming company said that “zero harm” means “zero tolerance” to even minor safety breaches and provided examples of how he has conveyed this to drivers.

This practice, managers take responsibility and show leadership in making safety a clear priority, is difficult to measure through safety management audits. Measurement of responsibility and leadership is more complex than the other practices covered by this research. Asking employees and drivers provides some insights but their responses may be biased. There are examples in the safety management literature from which to base tools to measure aspects of safety climate (Cox and Cheyne, 2000; Flin et al., 2000; Huang et al., 2013a; Huang et al., 2013b; Williamson et al., 1997; Zohar, 1980). A safety climate survey of seafarers found relationships between safety policy and supervisory behaviour, and in turn the perception of supervisory safety behaviour, positively related to seafarers’ safer behaviour (Lu and Tsai, 2010). This suggests that efficacy of safety management leadership can be measured by the extent to which all potential safety influencers in an organisation adopt the same level of safety commitment in their practices, and are seen to do so. A robust approach to measuring management responsibility-taking and leadership might be to select some relevant, objective indicators of these practices.

6.6 How the evidence-based safety management practices combine and interact

The management practices in SMS do not exist in isolation, nor are they static (Le Coze, 2008). As shown in Chapter 2, the literature is rich with evidence of the need to see organisational systems as dynamic organisms that have interacting parts (Hale et al., 1997; Mooren et al., 2009a; Rasmussen, 1997). This research identified characteristics and practices that distinguished lower- and higher-claiming companies but the combinations of other practices found to exist in both lower and higher claimers may have influenced the effects of each, or some of the practices with which they co-exist. Huang et al (2006b) argued that the mediating effects of each safety control element mean that the stronger each element is, the stronger the effectiveness of other elements will be. For example, they found that employee belief that they can control their safety behaviour, in turn improves safety climate and reduces self-reported injury incidents. This means that employee belief that they can control work related injury risk strengthens the belief that the safe way to do things is the way things are done in the company, and, in turn, this influences

172

safety outcomes. Similarly Al-Refaie (2013) showed that management commitment moderates the effectiveness of incentives, safety reporting and empowerment of employees, and that continual improvement and teamwork influence safety awareness. Safety leadership influences the effectiveness of a range of safety management practices and worker behaviours at all levels of an organisation. Based on a survey of drivers, dispatchers and safety directors of 116 US trucking companies, Arboleda et al (2003) argue that management commitment, opportunity for safety input and safety training influence perceptions of safety culture.

Moreover, the existence of other safety management practices influences the effectiveness of different safety management practices. For example, some payment systems have been linked with poor truck maintenance practices (Thompson and Stevenson, 2014) where there is an incentive to keep trucks on the road, skipping maintenance checks, in effort to optimise income.

Further research is needed to test the specific influences, and directions of influences, of the 14 practices in the proposed safety management system. Based on the weight of the safety literature it is very likely that moderating influences among some or all of the practices exist, particularly the moderating effect of the safety culture management practices on some or all of the other practices.

The dynamic system is embodied in the proposed SMS, summarised in Figure 6.2, as three interacting groups of practices or cogs: risk assessment and management, driver risk management and safety culture management. All three cogs are interconnected. Good safety culture management is conceived as a process that continuously influences and reflects vigilance in risk management and safe driver behaviour (DeJoy et al., 2010; Lu and Yang, 2010). Based on her research into workplace safety management, Makin (2009) argues that risk assessment and management are continuous processes of identifying potential safety hazards and influencing safety culture and worker safety practices. The management of driver related risk is another crucial cog in the SMS. Based on a study of a large coal-mining corporation using structural modelling techniques Cui et al (2013) argue that management commitment influences employee beliefs, and that employee beliefs influence safety involvement. Safety-focused drivers play an important role in continually identifying risks and contributing to risk management solutions and in turn can influence the other cogs in the system.

In this overall view of the SMS proposed for heavy vehicle transport, each of the three cogs contain components that interact with other components thus strengthening safety management effectiveness. For example, encouraging driver input in WHS decision-making makes driver incentive programs more effective. Similarly, site and job risk assessments can improve the effectiveness of safety training.

173

1. Managers encourage driver input into WHS decision-making. 2. Managers take responsibility and

safety features on trucks, is provided. 2. Maintenance and pre-trip vehicle checks ensure that trucks are in safe conditions for all bips. 3. Monitor fatigue management practices. 4. Route risk a&&ll&&llllllll, including lnllic Clllllilianl ..dane far ..

Figure 6.2 Model of an integrated safety management system (SMS) for heavy vehicle transport

In this model, optimal safety management involves these interlocking cogs with interacting components. The model also envisages safety culture management as a continual process of demonstrating safety leadership and expectations that in turn encourage vigilant risk assessment and amelioration. Other research has demonstrated that the strength of safety culture and communication is likely to intensify the effectiveness of driver risk management (AI-Refaie, 2013). Moreover, AI-Raife concluded that the management of driver risk assists to build a stronger safety culture and enhance risk perceptions and safety innovations. Based on research find ings in Portuguese chemical companies Silva and Lima (2005) argued that an ongoing risk assessment process helps to build strong safety awareness by managers, drivers and others in the organisation, as well as contributing to learning and development that, in turn, strengthens safety culture. Finally, the findings of safety management research (Leveson, 2004) in the

174

aerospace industry suggest that risk assessment and management, driver risk management and safety culture management practices are continuously changing and interacting with one another.

The scientific research literature, described in this section, provides evidence that there are synergistic effects of combining safety management practices. The yet-untested relationships between safety outcomes and combined safety management practices could provide further evidence that the SMS would operate as an integrated system with interacting elements.

6.7 General discussion

The genesis of this thesis work was recognition that despite high levels of safety risk in the heavy vehicle transport sector, very little research had been done to test the efficacy of safety management practices in either in this sector or any other sector.

The research for this thesis comprised three distinct studies that built an evidence base for the development of a safety management system (SMS) for heavy vehicle road transport operations. The method used in the identification of important safety management practices was to distil from the scientific literature positive relationships between safety management practices and safety outcomes, then to work backwards looking at good and poorer safety performers and seeing what combinations of safety management practices distinguished between them in a sample of heavy vehicle truck operators. The synthesis of knowledge gained through the studies together with an examination of the road freight transport sector enabled development of an SMS particularly suited for this industry. Each of the fourteen practices of the proposed SMS have been justified by original research (Study 2 and Study 3), together with evidence from the scientific safety literature. The unexpected findings, where what were thought to be good safety management practices were found to be more common amongst higher insurance claimers than lower claimers, suggest that practices like, having safety accreditations, safety policies and training, are not sufficient for the achievement of good safety outcomes – and indeed could mislead regulators and customers that the company has good safety outcomes.

The proposed SMS recognises the interplay of applying the safety management practices together interactively. In other words it is hypothesised that the synergistic effects of risk assessment management, driver risk management and safety culture management elements working together mean that comprehensive implementation of the SMS is likely to be more effective than implementing one or several of the elements alone. However, this was not tested in this research. Testing the SMS on a poorer (safety) performing company to see if its crash outcomes improve was beyond the scope of this thesis. It will be up to other researchers in a future research project to test the SMS. The benefits of this research have been the identification of a set of safety management practices that were shown to distinguish between lower and higher-claiming heavy vehicle transport operators. This provides a more targeted approach to safety management improvements.

175

6.7.1 Limitations Limitations relating to data collection and analysis in Studies 2 and 3 have been discussed in Chapters 4 and 5. There were significant challenges in recruiting participant companies for the study, resulting in the inability to rely on formal tests of statistical significance for confirming survey results. However, tests for effects sizes (Olivier and Bell, 2013) to provide guidance on the robustness of the survey findings were carried out.

Another limitation to the survey was that, like most similar studies, it relied on self-reported data provided by managers about their safety management. This was recognised as a potential problem from the outset. In part for this reason the study included a validation process following the survey in the form of an in- depth investigation into a sample of survey participants. That investigation specifically sought to identify evidence to validate the accuracy of the answers provided in the survey.

Due to the small survey sample size, for many purposes it was not possible to analyse the survey data within size groupings (small and large fleet operators). However, as discussed in Chapter 5, the in-depth investigations found that while smaller companies had less detailed formal systems generally, the more distinguishable safety management characteristics were the ways in which drivers perceived the importance that management placed on safety, regardless of company size. Otherwise, the differences found were more attributable to claim rate performance than size of company.

It is also noted that as the participants were all companies that were operating only in Australia, it is not known if the findings and specific outputs of the study will apply in other countries. The study has particular currency in Australia where regulators and industry are grappling with the transport regulations that do not require auditable safety management systems as are required in other countries, such as the USA, including regulation of ‘chain of responsibility’ whereby operators are not told how they should implement safety management practices, but instead must show a ‘duty of care.’ However, the study can also assist to inform the efficacy of practices prescribed in existing auditable safety management systems or safety management regulatory requirements. The nature and scope of this type of study, where there were many variables to consider within a complex system, did not permit a detailed examination of each of the safety management practices and the effects of applying these practices. This was deliberate, as the unit of study was the company and not an evaluation of management action or company practices within a single company.

6.7.2 Empirical testing of the SMS The research presented in this thesis has led to the development of an evidence-based framework of practices of an SMS this is likely to be effective in improving safety and crash outcomes. Currently, a project to empirically test the safety management system has been commenced by the University of New South Wales. The SMS will be trialled in companies that operate 10-50 heavy vehicles for transport of goods, equipment and/or other materials under hire and reward conditions. These types of companies

176

were chosen to produce a relatively homogenous sample for the evaluation. Very small companies that might have limited resources to undertake the study and very large companies that might have difficulty implementing change within the study timeframes will not be recruited. The intervention will be applied in lower safety performing companies, using the safety measure of insurance claims per truck. This criterion is similar to that used in the study presented in this thesis.

The trial study will be an intervention-control group design conducted over two years. Baseline measures of practices and outcomes will be taken of all the companies during a ‘pre-intervention’ (baseline) period at the start of the first year. The companies will then be randomly assigned to an Intervention or a ‘wait’ control group, with the constraint that there is a similar mix of larger and smaller companies in each group. The plan is for the first intervention group to receive the intervention during the first year. At the end of the first year, practices and outcomes will be measured again in all companies. Half of the control group will then be selected randomly to receive the intervention during the second year. At the end of year two, all the companies will be measured again. This design permits an analysis of both short and longer-term effects of the intervention. At the end of the study the remaining control group companies will be given the intervention. The main components of the intervention are an initial assessment of each company’s practices relative to the set of characteristics identified in the earlier studies, structured assistance in identifying improved practices suitable for each company, development and monitoring of an implementation schedule, and access to ongoing advice about implementation. Using the evidence-based safety management system, further trial research in the transport sector, as well as other industry sectors, will be encouraged.

6.7.3 Recommendations for further research This research found that a number of safety management items were not confirmed as characteristics that distinguish between lower and higher truck insurance claimers because the managers did not provide sufficient evidence in Study 3 to validate their survey answers. There could be benefit in testing these characteristics in future studies. The findings that were unexpected, based on a review of findings from prior research, particularly the finding that higher claimers were more likely to have fatigue risk management polices and training, certainly warrant further investigation, although one interpretation of this is that these practices are not important safety management practices on their own. But this interpretation should be tested to learn more about what is particularly important in managing fatigue risk in truck driving.

Additionally research to test and refine relationships between elements in the model is recommended. The big challenge is to find methods to research relationships in a non-linear dynamic model. For example, the provision of monetary incentives to drivers might strengthen the site and job risk assessment element of the SMS, which is also influenced by the extent to which managers encourage driver input into WHS decision-making, but these relationships need to be tested.

177

As injury events occur from a set of interacting causes, prevention may require sets of interacting safety management practices. In other words, the whole is greater than the sum of its parts – and choosing only one or two practices to implement is not likely to be as effective as choosing a system of interacting practices. This is especially the case for the dynamic and complex operations that characterise heavy vehicle transport operations. In this industry work practices do not generally involve only predictable, repetitive tasks. They are instead practices in which the changing environment and safety challenges demand an ability to make risk decisions quite frequently throughout the work process. For example, a driver must constantly select optimal driving speeds, and constantly decide whether s/he is mentally or physically fit enough to continue to make safe decisions. Further research into the specific work and decision-making processes undertaken by truck drivers would assist to gain insights into the risk processes involved in truck driving. A naturalistic driving24 study involving truck drivers would be helpful.

6.7.4 Implications for industry, regulators and insurers Notwithstanding the limitations mentioned above, the development of an evidence-based safety management system suitable for use by heavy vehicle transport operators potentially could assist large and small trucking companies and others that operate heavy trucks to reduce safety risks by implementing the system. The results of this research could be presented in public forums, and companies operating heavy vehicles may wish to consider the possible uses of the findings.

If the empirical testing of the SMS, described above, finds that implementing the SMS does improve safety outcomes, industry and government safety management schemes could then be refined and/or elements of existing schemes supported. Ultimately, this safety management system, made freely available, could assist heavy vehicle transport operators to comply with their chain of responsibility obligations. This in turn should assist regulators to know what to look for to determine if companies are meeting their duty of care responsibilities. Similarly, insurance companies that provide truck insurance may be in a better position to determine risk propensities of current and future policyholders and to suggest interventions to improve safety performance and thus reward companies that implement such a system with lower premiums.

6.7.5 Changing the paradigm This study set out to develop a safety management system through a research process. Most would say that the bottom line for safety interventions is absence or reduced incidence of harmful events (Nilsen et al., 2004). The approach used in this research was to start from outcomes (comparative insurance claim rates) and work back to the characteristics that distinguish good and poorer safety performing companies in order to design interventions that should improve safety outcomes. It aimed to find characteristics that distinguished between companies with better safety performance and companies with poorer safety performance. It did not seek to test the specific effects of particular management practices, but rather the approach was to find as many elements of safety management that characterised good safety performers

24 Naturalistic driving studies involve instrumenting vehicles with monitoring equipment including video cameras and other data collection devices to enable an examination of driving behaviours.

178

and did not characterise poor safety performers. Looking at safety management more holistically allows the identification of sets of important safety management practices.

This contrasts with the traditional road safety dissection of contributing injury factors when crashes occur and then planning individual countermeasures to address the individual contributing factors (Salmon et al., 2012). Road safety research has taught us that combinations of countermeasures such as public education and enforcement are more effective than doing one without the other (Elliott, 1993; Lewis et al., 2007). However, by and large, interventions are planned and implemented in a silo-fashion to address road environment factors, vehicle and equipment factors and human factors as separate problems. That is, traditionally, road safety researchers and practitioners have approached road injury hazards as individual problems and have in general failed to acknowledge the dynamic interactions among them.

Taking a systems view, road injury hazards can be understood as a set of interacting variables that require interdependent actions to respond effectively to these risks instead of the traditional approach of a group of individual problems with sets of single interventions. This relatively new systems approach to road safety reflects a greater understanding of the complex and integrated nature of human interaction with the environments in which they operate within (Salmon et al., 2012). However, the systems approach has not been fully applied in road safety.

There are parallels between public road safety and workplace safety but up until now, the two fields have taken different approaches to analysing and managing risk. Just as road rules without enforcement are not as effective in encouraging safe behaviour so too are safety policies without consistent corrective actions in the workplace from WHS regulators. Reason (2000) speaks of active failures and latent conditions (caused by human decisions prior to the event). In this way, injury events are understood as resulting from inadequate injury defences in a system or process. Latent conditions can be addressed before proximal risk factors have a chance to manifest. For example, we know that unsafe driving speed is a major factor in fatal crashes, but we are only starting to understand what can be done at a systemic level to change the conditions that encourage speeding. The decision by a truck driver to drive at unsafe speeds, is in part, influenced by organisational pressures and financial conditions and decisions – both at a personal and at a company level. Indeed, this study has reinforced the evidence that payment for work done influences safety outcomes in the trucking industry. In this case, the choice of driver remuneration method is a systemic response to the risk of speeding.

The ‘safe system’ principle adopted to underpin road safety strategies in Australia and in other jurisdictions implies that ‘systems thinking’ is starting to gain attention by road safety researchers and practitioners. The idea is for the road traffic system to be designed and managed to eliminate inherent injury risks, largely by making crashes more survivable. The current road safety thinking is that it is not possible to perfect human road users such that crashes can all be avoided through consistent correct decision- making. In effect, the road safety systems approach is currently less concerned about preventing crashes,

179 than protecting road users from being harmed in the event of a crash. The prevailing objective of the New Car Assessment Program (NCAP) is to ensure vehicle occupants are cushioned (by seatbelts and airbags) to prevent them from harmful impacts within the vehicle when it crashes. NCAP is only now beginning to test the efficacy of crash avoidance systems as well. Similarly the international Road Assessment Program has focussed more on developing roads and roadsides that are forgiving of human error. The idea is that human error is assumed to be a given condition. The mixed results in the literature about whether driver training can reduce crashes reinforces the notion that skilled and experienced drivers are unlikely to become less error-prone through educational measures. Therefore the behavioural countermeasures that are used by road safety authorities are largely focused on punitive actions for traffic violations. Developing and maintaining a road safety culture is rarely even discussed in this field, unlike the WHS field.

In workplace safety, safety culture and system safety is more fully addressed with accident prevention as well as mitigation strategies. These strategies involve efforts to reduce the likelihood of human error as well as safeguarding the work environment and work process such that workers do not get harmed in accidents by managing injury risks in the work system. This study has attempted to show how a systems approach can be applied to road safety where government authorities, as well as companies, are the system managers ultimately responsible for ensuring safety within the system. Moreover, if an incident occurs these authorities, companies and managers must ensure that the incident is not repeated.

The study may provide a new impetus to shift the road safety paradigm to a systems analytical approach.

6.8 Conclusions

The overall aims of this thesis were met. Through a three-staged research process the study has identified the distinguishing characteristics of heavy vehicle transport operating companies with good safety records and those with poorer safety records, and developed a safety management system suitable for heavy vehicle transport operations. This study has done all it can to ensure that the evidence on the components of a successful SMS for heavy transport is as strong as possible. This was done by:

Basing identification of practices on the solid foundation of previous research; Identifying practices that distinguish lower and higher performing companies; Checking the validity of these identified practices; and Identifying some practices that might not actually contribute to better safety despite previous research that suggests they should be included in an SMS.

There is strong evidence that fully implementing the 14 safety management practices of the proposed safety management system should improve safety outcomes in companies that operate heavy trucks. The effectiveness of implementing some, but not all, of the recommendations may reduce the ability to achieve

180

optimal outcomes as many of the elements, and groups of characteristics, can influence the effects from other elements.

Developing an evidence-based safety management system advances the field of occupational safety, particularly as it relates to work related road safety. It provides a new model for exploration, testing and refinement. This ultimately is likely to result in fewer work related road injuries in the heavy vehicle transport sector, as well as translation and application in other work related road safety contexts.

While no one has yet applied an SMS approach to road safety, and demonstrated that the method can reduce crashes and associated road trauma, there is reason to believe that this would be worthwhile. Now there is an evidence-based SMS tailored to the heavy trucking industry to trial. The planned study involving the implementation and evaluation of this SMS in companies with poor safety outcomes is expected to further confirm the effectiveness of the new SMS.

Finally, this work has challenged traditional approaches to road safety analysis and intervention planning in fundamental ways. Analysis of this study’s findings indicates that a more integrated and dynamic model is needed, to apply to road safety research and countermeasure planning, using a systems approach.

181

References

Abad, J., Lafuente, E., Vilajosana, J., 2013. An assessment of the OHSAS 18001 certification process: Objective drivers and consequences on safety performance and labour productivity. Safety Science 60, 47-56.

ACIL_Tasman, 2003. Freight rates and safety performance in the road freight industry. National Transport Commission.

ACIL_Tasman, 2004. Trucking - driving Australia's growth and prosperity. Australian Trucking Association.

Acton, J., 2012. Road safety remunerations tribunal. Road Safety Remuneration Tribunal,, Australia, p. Web page with information about the Tribunal.

Adams-Guppy, J., Guppy, A., 2003. Truck driver fatigue risk assessment and management: A multinational survey. Ergonomics 46, 763-779.

Al-Refaie, A., 2013. Factors affect companies’ safety performance in Jordan using structural equation modeling. Safety Science 57, 169-178.

American Transport Research Institute, 2008. A technical analysis of driver training impacts on safety. ATRI, Arlington, Virginia.

Apostolopoulos, Y., Sönmez, S., Shattell, M., Belzer, M., 2012. Environmental determinants of obesity‐ associated morbidity risks for truckers. International Journal of Workplace Health Management 5, 120-138.

Arboleda, A., Morrow, P., Crum, M., Shelley, L., Mack, C., 2003. Management practices as antecedents of safety culture within the trucking industry: Similarities and differences by hierarchical level. Journal of Safety Research 34, 189-197.

ATC, 2004. National road safety initiatives.

ATN, 2012. CoR investigation launched after police raid lennons transport, Australian Transport Network,.

ATN, 2015. CoR resources the focus of Victoria Police scrutiny, Australian Transport News.

Australian Industry Group, 2012. The outlook for the transport sector.

Australian Logistics Council, National Logistics Safety Code.

Australian Logistics Council, 2015. ALC 2015 yearbook future freight networks, Canberra.

Australian Transport Safety Bureau, 2008. Statistical update progress report presented to National Road Safety Strategy Panel, Sydney.

Australian Trucking Association, 1991. Trucksafe.

Australian Trucking Association, 2014. Trucksafe operator business rules and code of conduct. Australian Trucking Association.

AustRoads, 2010. Guide to road design part 6: Roadside design, safety and barriers. AustRoads.

Baas, P., Charlton, S., Bastin, G., 2000. Survey of New Zealand truck driver fatigue and fitness for duty. Transportation Research Part F: Traffic Psychology and Behaviour 3, 185-193.

182

Baas, P., Taramoeroa, N., 2008. Analysis of the safety benefits of heavy vehicle accreditation schemes. Austroads, Sydney.

Backman, A., Järvinen, E., 1983. Turnover of professional drivers. Scandinavian Journal of Work, Environment & Health 9, 36-41.

Baker, S., Edwards, R., 2012. How many qualitative interviews is enough. National Centre for Research Methods,, United Kingdom.

Banks, T., 2008. An investigation into how work-related road safety can be enhanced, Centre for Accident Research & Road Safety - Qld (CARRS-Q). Queensland University of Technology.

Belman, D., Monaco, K., 2001. The effects of deregulation, de-unionisation, technology, and human capital on the work and work lives of truck drivers. Industrial and Labor Relations Review 54, 502-524.

Belzer, M., 2000. Sweatshops on wheels: Winners and losers in trucking deregulation. Oxford University Press.

Belzer, M., Rodriguez, D., Sedo, S., 2002. Paying for safety: An economic analysis of the effect of compensation on truck driver safety. Science Applications International Corporation, Federal Motor Carrier Safety Administration.

Besnard, D., Hollnagel, E., 2014. I want to believe: Some myths about the management of industrial safety. Cognition, technology and work 16, 13-23.

Bezwada, N., 2010. Characteristics and contributory causes associated with fatal large truck crashes, Department of Civil Engineering. Kansas State University, Manhattan.

BITRE, 2011. Truck productivity: Sources, trends and future prospects, Canberra, ACT.

BITRE, 2015a. International road safety comparisons 2013.

BITRE, 2015b. Road trauma involving heavy vehicles 2014 statistical summary, Canberra.

Bjerkan, A., 2010. Health, environment, safety culture and climate - analysing the relationships to occupational accidents. Journal of Risk Research 13, 445-477.

Blower, D., Green, P., Matteson, A., 2010. Condition of trucks and truck crash involvement. Transportation Research Record: Journal of the Transportation Research Board 2194, 21-28.

Bottani, E., Monica, L., Vignali, G., 2009. Safety management systems: Performance differences between adopters and non-adopters. Safety Science 47, 155-162.

Boufous, S., Williamson, A., 2006. Work-related road fatalities: A record linkage study. Accid Anal Prev 38, 14-21.

Bradley, A., 2014. Case study - Nestle, Driving for Better Business. RoadSafe.

Braver, E., Preusser, C., Preusser, D., Baum, H., Beilock, R., Ulmer, R., 1992. Who violates work hours rules? A survey of tractor-trailer drivers. Insurance Institute for Highway Safety, Arlington, Virginia.

British Telecom, 2010. Case study - BT, Driving for Better Business. RoadSafe.

Brodie, L., Bugeja, L., Elias, I., 2009. Heavy vehicle driver fatalities: Learnings from fatal road crash involvement in Victoria. Accid Anal Prev 41, 557-564. 183

Brommelsiek, W., Tinsley, K., 1996. Environmental and safety audits - a process to improve business performance, In: Engineers, S.o.P. (Ed.), International Conference on Health, Safety & Environment. Society of Petroleum Engineers, New Orleans, Louisiana, pp. 399-403.

Brooks, C., 2002. Speed and heavy vehicle safety, National Heavy Vehicle Safety Seminar. National Transport Commission, Melbourne.

Brown, R., Holmes, H., 1986. The use of a factor-analytic procedure for assessing the validity of an employee safety climate model. Accident Analysis & Prevention 18, 455-470.

Bruning, E., 1989. The relationship between profitability and safety performance in trucking firms. Transportation Journal 28, 40-49.

BSI, 2014. As/nz 4801:2001 safety management systems (SMS) self-assessment checklist.

Burgess-Limerick, R., Bowen-Rotsaert, D., 2002. Fatigue management program pilot evaluation: Phase 2 wave 3 report. Queensland: Global Institute of Learning and Development Consortium, 48.

Cantor, D., Corsi, T., Grimm, C., Ozpolat, K., 2010. A driver focused truck crash prediction model. Transportation Research Part E: Logistics and Transportation Review 46, 683-692.

CASA, 2011. Australia's state safety program. Australian Government.

Cassini, P., 1998. Road transportation of dangerous goods: Quantitative risk assessment and route comparison. Journal of Hazardous Materials 61, 133-138.

Centre for International Economics, 2011. Benefit-cost analysis for the regulation impact statement on the heavy vehicle National law, Canberra.

Chen, F., Chen, C., 2014. Measuring the effects of safety management system practices, morality leadership and self-efficacy on pilots’ safety behaviors: Safety motivation as a mediator. Safety Science 62, 376-385.

Chen, F., Chen, S., 2011. Reliability-based assessment of vehicle safety in adverse driving conditions. Transportation Research Part C: Emerging Technologies 19, 156-168.

Choudhry, R., Fang, D., Mohamed, S., 2007. The nature of safety culture: A survey of the state-of-the-art. Safety Science 45, 993-1012.

Cigularov, K., Chen, P., Rosecrance, J., 2010. The effects of error management climate and safety communication on safety: A multi-level study. Accident Analysis & Prevention 42, 1498-1506.

Cohen, J., 1992. A power primer. Psychological Bulletin 112, 155-159.

Comcare, 2011. Guidance for officers in exercising due diligence. Australian Goverment.

Corsi, T., Barnard, R., Gibney, J., 2002. Motor carrier industry profile: Linkages between financial and safety performance among carriers in major industry segments. Department of Transportation, Washington, DC., pp. 1-84.

Corsi, T., Grimm, C., Cantor, D., Sienicki, D., 2012. Safety performance differences between unionized and non-union motor carriers. Transportation Research Part E: Logistics and Transportation Review 48, 807-816.

Cox, S., Cheyne, A., 2000. Assessing safety culture in offshore environments. Safety Science 34, 111-129.

184

Coyle, I., Sleeman, S., Adams, N., 1995. Safety climate. Journal of Safety Research 26, 247-254.

Crum, M., Morrow, P., 2002. The influence of carrier scheduling practices on truck driver fatigue. Transportation Journal 42, 20-41.

Cui, L., Fan, D., Fu, G., Zhu, C., 2013. An integrated model of organizational safety behavior. Journal of Safety Research 45, 37-46.

Dahl, Ø., Fenstad, J., Kongsvik, T., 2013. Antecedents of safety-compliant behaviour on offshore service vessels: A multi-factorial approach. Maritime Policy & Management 41, 20-41.

Darby, P., Murray, W., Raeside, R., 2009. Applying online fleet driver assessment to help identify, target and reduce occupational road safety risks. Safety Science 47, 436-442.

De Croon, E.M., Blonk, R.W., De Zwart, B.C., Frings-Dresen, M.H., Broersen, J.P., 2002. Job stress, fatigue, and job dissatisfaction in dutch lorry drivers: Towards an occupation specific model of job demands and control. Occupational and Environmental Medicine 59, 356-361. de Pont, J., 2005. Assessing heavy truck safety in Tasmania. Department of Infrastructure, Energy and Resources, Hobart.

Dean, J., 1947, reprinted 2007. Murder most foul. George Allen & Unworth, London.

Dedobbeleer, N., Beland, F., 1991. A safety climate measure for construction sites. Journal of Safety Research 22, 97-103.

DeJoy, D., Della, L., Vandenberg, R., Wilson, M., 2010. Making work safer: Testing a model of social exchange and safety management. Journal of Safety Research 41, 163-171.

Dekker, S., 2011. Drift into failure: From hunting broken components to understanding complex systems. Ashgate, Aldershot, UK.

Department of Transportation U.S., 2006. Report to congress on the large truck crash causation study. National Technical Information Service, Virginia.

Dillman, D., 2009. Internet, mail, and mixed-mode surveys: The tailored design method. Wiley, New York.

Dingus, T., Neale, V., Klauer, S., Petersen, A., Carroll, R., 2006. The development of a naturalistic data collection system to perform critical incident analysis: An investigation of safety and fatigue issues in long- haul trucking. Accident Analysis & Prevention 38, 1127-1136.

Dixon-Woods, M., Agarwal, S., Jones, D., Young, B., Sutton, A., 2005. Synthesising qualitative and quantitative evidence: A review of possible methods. Journal of Health Services Research & Policy 10, 45- 53.

Driver, P., 2015. Partnership audit review, 2015 ALC Supply Chain Safety & Compliance Summit. Australian Logistics Council, Sydney.

Dubens, E., Murray, W., 2009. The status of occupational road safety globally: Capsule presentation of conference white paper First International Conference on Road Safety at Work. Virtual Risk Manager, Washington DC.

Eakin, J., Champoux, D., MacEachen, E., 2010. Health and safety in small workplaces: Refocusing upstream. Canadian Journal of Public Health March/April, S29-S33.

185

Edkins, G., 1998. The indicate safety program: A method to proactively improve airline safety performance. Safety Science 30, 275-295.

Elkington, J., Stevenson, M., 2013. The heavy vehicle study. Curtin University.

Elliott, B., 1993. Road safety mass media campaigns: A meta-analysis. Federal Office of Road Safety,, Canberra.

Elvik, R., 2009. An exploratory analysis of models for estimating the combined effects of road safety measures. Accident Analysis and Prevention 41, 876-880.

Employment and Social Development Canada, 2014. Occupational injuries amongst Canadian federal jurisdiction employers, 2007-2011. Government of Canada, Ottawa.

Fernandez-Muniz, B., Montes-Peon, J., Vazquez-Ordas, C., 2007a. Safety culture: Analysis of the causal relationships between its key dimensions. Journal of Safety Research 38, 627-641.

Fernandez-Muniz, B., Montes-Peon, J., Vazquez-Ordas, C., 2009. Relation between occupational safety management and firm performance. Safety Science 47, 980-991.

Fernández-Muñiz, B., Montes-Peón, J., Vázquez-Ordás, C., 2014. Safety leadership, risk management and safety performance in spanish firms. Safety Science 70, 295-307.

Fernandez-Muniz, B., Montes-Peon, J., Vazquez-Ordas, C.J., 2007b. Safety management system: Development and validation of a multidimensional scale. Journal of Loss Prevention in the Process Industries 20, 52-68.

Ferrier Hodgeson, 2014. Transport and logistics insights: The road ahead.

Feyer, A., Williamson, A., 1995. Work and rest in the long-distance road transport industry in Australia. Work & Stress 9, 198-205.

Feyer, A., Williamson, A., Friswell, R., Sadural, S., 2002. Driver fatigue: A survey of long distance transport companies in Australia. Australian Transport Safety Bureau, Canberra.

Flin, R., Mearns, K., O'Connor, P., Bryden, R., 2000. Measuring safety climate: Identifying the common features. Safety Science 34, 177-192.

FMCSA, 2006. Report to Congress on the large truck crash causation study.

Fogarty, G., Shaw, A., 2010. Safety climate and the theory of planned behavior: Towards the prediction of unsafe behavior. Accident Analysis & Prevention 42, 1455-1459.

Frazier, C., Ludwig, T., Whitaker, B., Roberts, D., 2013. A hierarchical factor analysis of a safety culture survey. Journal of Safety Research 45, 15-28.

Friswell, R., 2013. Fatigue and occupational safety in short haul, light transport, PhD Thesis School of Public Health and Community Medicine. University of New South Wales.

Friswell, R., Williamson, A., 2010. Work characteristics associated with injury among light/short haul transport drivers. Accid Anal Prev 42, 2068-2074.

Gander, P., Marshall, N., Bolger, W., Girling, I., 2005. An evaluation of driver training as a fatigue countermeasure. Transportation Research Part F: Traffic Psychology and Behaviour 8, 47-58.

Gardner, B., 2013. Austrans results spark CoR investigations in NSW, Australian Transport Network,.

186

Geldart, S., Smith, C., Shannon, H., Lohfeld, L., 2010. Organizational practices and workplace health and safety: A cross-sectional study in manufacturing companies. Safety Science 48, 562-569.

Golob, T., Hensher, D., 1994. Driver behavior of long distance truck drivers: The effects of schedule compliance on drug use and speeding citations. Institute of Transport Studies, University of California, Irvine.

Grayson, B., Helman, S., 2011. Work-related road safety: A systematic review of the literature. IOSH, Leicestershire, England.

Gregersen, N.-P., Brehmer, B., Moren, B., 1996. Road safety improvement in large companies. An experimental comparison of different measures. Accident Analysis & Prevention 28, 297-306.

Grimm, P., 2010. Social desirability bias, Wiley international encyclopedia of marketing. John Wiley & Sons, Ltd.

Grote, G., Weichbrodt, J., Gunter, H., Zala-Mezo, E., Kunzler, B., 2009. Coordination in high-risk organizations: The need for flexible routines. Cognitive Technical Work 11, 17-27.

Grzebieta, R., Mooren, L., Job, S., 2013. Introduction (or reintroduction) to the safe system approach. Transportation Research Circular E-C172: Road Safety Design and Devices.

Grzebieta, R., Rechnitzer, G., 2001. Crashworthy systems - paradigm shift in road safety design (part ll). IEAust 7.

Guest, M., Boggess, M., Duke, J., 2014. Age related annual crash incidence rate ratios in professional drivers of heavy goods vehicles. Transportation Research Part A: Policy and Practice 65, 1-8.

Guldenmund, F., 2000. The nature of safety culture: A review of theory and research. Safety Science 34, 215-257.

Haddon, W., 1968. The changing approach to the epidemiology, prevention, and amelioration of trauma: The transition to approaches etiologically rather than descriptively based. American Journal of Public Health 58, 1431-1438.

Haddon, W., 1980. Advances in the epidemiology of injuries as a basis for public policies. Landmarks in American Epidemiology 95, 411-421.

Hale, A., Heming, B., Carthey, J., Kirwan, B., 1997. Modelling of safety management systems. Safety Science 26, 121-140.

Hale, A., Hovden, J., 1998. Management and culture: The third age of safety. A review of approaches to organizational aspects of safety, health and environment, In: Feyer, A.-M., Williamson, A. (Eds.), Occupational injury: Risk, prevention and intervention. Taylor & Francis, London, pp. 129-166.

Hanowski, R., Hickman, J., Olsen, R., Bocanegra, J., 2009. Evaluating the 2003 revised hours-of-service regulations for truck drivers: The impact of time-on-task on critical incident risk. Accident Analysis & Prevention 41, 268–275.

Heinrich, H., 1931. Industrial accident prevention: A scientific approach. McGraw-Hill, New York.

Helmreich, R., 2000. On error management: Lessons from aviation. BMJ 320, 781-785.

187

Hensher, D., Battellino, H., 1990. Long-distance trucking: Why do truckies speed?, Australasian Transport Research Forum, pp. 537-554.

Hensher, D., Battellino, H., Gee, J., Daniels, R., 1991. Long distance truck drivers on-road performance and economic reward. University of Sydney,, Washington DC.

Hensher, D., Daniels, R., Battellino, H., 1992. Safety and productivity in the long distance trucking industry, 16th ARRB Conference, Perth.

Higgins, J., Green, S.e., 2008. Cochrane handbook for systematic reviews of interventions. Wiley- Blackwell, West Sussex, England.

Hollnagel, E., 2002. Understanding accidents-from root causes to performance variability, Human factors and power plants, 2002. proceedings of the 2002 ieee 7th conference on. IEEE, pp. 1-1-6.

Hollnagel, E., 2008. Risk + barriers = safety? Safety Science 46, 221-229.

Hollnagel, E., 2014. Safety-I and safety-II: The past and future of safety management. Ashgate, Farnham.

Hollnagel, E., Woods, D., Leveson, N., 2006. Resilience engineering: Concepts and precepts. Ashgate.

Hollnagel, E., Woods, D., Leveson, N., 2007. Resilience engineering: Concepts and precepts. Ashgate, Farnham.

Hopkins, A., 2000. Lessons from longford: The esso gas plant explosion. CCH Australia Ltd.

Hopkins, A., 2006. Studying organisational cultures and their effects on safety. Safety Science 44, 875- 889.

Hovden, J., Albrechtsen, E., Herrera, I., 2010. Is there a need for new theories, models and approaches to occupational accident prevention? Safety Science 48, 950-956.

Huang, Y.-H., Ho, M., Smith, G., Chen, P., 2006a. Safety climate and self-reported injury: Assessing the mediating role of employee safety control. Accident Analysis & Prevention 38, 425-433.

Huang, Y.-H., Ho, M., Smith, G.S., Chen, P.Y., 2006b. Safety climate and self-reported injury: Assessing the mediating role of employee safety control. Accident Analysis & Prevention 38, 425-433.

Huang, Y.-H., Zohar, D., Robertson, M., Garabet, A., Lee, J., Murphy, L., 2013a. Development and validation of safety climate scales for lone workers using truck drivers as exemplar. Transportation Research Part F: Traffic Psychology and Behaviour 17, 5-19.

Huang, Y.-H., Zohar, D., Robertson, M., Garabet, A., Murphy, L., Lee, J., 2013b. Development and validation of safety climate scales for mobile remote workers using utility/electrical workers as exemplar. Accident Analysis & Prevention 59, 76-86.

Hudson, P., 2007. Implementing a safety culture in a major multi-national. Safety Science 45, 697-722.

Hughes, B., Newstead, S., Anund, A., Shu, C., Falkmer, T., 2015. A review of models relevant to road safety. Accident Analysis & Prevention 74, 250-270.

IBISWorld, Road freight transport: Competitive landscape.

James, P., Johnstone, R., Quinlan, M., Walters, D., 2007. Regulating supply chains to improve health and safety. Industrial Law Journal 36, 163-187.

188

Jansen, C., Dikranian, G., 2009. Heavy vehicle compliance survey 2009. NSW Roads and Traffic Authority, Sydney.

Jansen, C.D., G., 2009. Heavy vehicle compliance survey 2009. NSW Roads and Traffic Authority, Sydney.

Johansson, R., 2009. Vision Zero - implementing a policy for traffic safety. Safety Science 47, 826-831.

Johnston, I., 2010. Beyond “best practice” road safety thinking and systems management – a case for culture change research. Safety Science 48, 1175-1181.

Johnstone, R., Mayhew, C., Quinlan, M., 2001. Outsourcing risk? The regulation of occupational health and safety where subcontractors are employed. Comparative Labor Law & Policy Journal 22.

Joy, J., 2004. Occupational safety risk management in Australian mining. Occupational medicine 54, 311- 315.

Kapp, E., 2012. The influence of supervisor leadership practices and perceived group safety climate on employee safety performance. Safety Science 50, 1119-1124.

Kath, L., Magley, V., Marmet, M., 2010. The role of organizational trust in safety climate's influence on organizational outcomes. Accident Analysis & Prevention 42, 1488-1497.

Kemp, E., Kopp, S., Kemp, E., 2013. Six days on the road: Will I make it home safely tonight? Examining attitudes toward commercial transportation regulation and safety. The International Journal of Logistics Management 24, 210-229.

Kines, P., Andersen, D., Andersen, L., Nielsen, K., Pedersen, L., 2013. Improving safety in small enterprises through an integrated safety management intervention. Journal of Safety Research 44, 87-95.

Kirwan, B., 1998. Safety management assessment and task analysis: A missing link?, In: Hale, A., Baram, M. (Ed.), Safety management: The challenge of change. Elsevier, Oxford.

Kjellén, U., 1984. The deviation concept in occupational accident control—I. Accident Analysis & Prevention 16, 289-306.

Kjellén, U., 2000. Prevention of accidents through experience feedback. CRC Press.

Kjellen, U., Larsson, T., 1981. Investigating accidents and reducing risks — a dynamic approach. Journal of Occupational Accidents 3, 129-140.

Knestaut, A., 1997. Fatalities and injuries among truck and taxicab drivers, Compensation and Working Conditions, Washington DC.

Knipling, R., Hickman, J., Bergoffen, G., 2003. Effective commercial truck and bus safety management techniques: A synthesis of safety practice, In: Board, T.R. (Ed.), Commercial truck and bus safety synthesis program. Federal Motor Carrier Safety Administration, Washington DC.

Knipling, R., Nelson, K., Bergoffen, G., Burks, S., 2011. Safety management in small motor carriers: A synthesis of safety practice, CTBSSP Synthesis 22, Washington DC.

Langwieder, K., Gwehenberger, J., Hummel, T., Bende, J., 2001. Benefit potential of ESP in real accident situations involving cars and trucks, 18th ESV Conference, Nagoya, Japan.

189

Lau, S.W., 1995. Truck travel surveys: A review of the literature and state-of-the-art. Metropolitan Transport Commission, Oakland, California.

Lawson, C., Riis, A.-E., 2001. We're really asking for it: Using surveys to engage the freight community. Transportation Research Record 1763, 13-19.

Le Coze, J.-C., 2008. Disasters and organisations: From lessons learnt to theorising. Safety Science 46, 132-149.

Leveson, N., 2004. A new accident model for engineering safer systems. Safety Science 42, 237-270.

Leveson, N., Dulac, N., Zipkin, D., Cutcher-Gershenfeld, J., Caroll, J., Barrett, B., 2007. Engineering resilience into safety-critical systems, In: Hollnagel, E., Woods, D., and Leveson, N. (Ed.), Resilience engineering: Concepts and precepts. Ashgate, Farnham, pp. 95-123.

Lewis, I., Watson, B., Tay, R., White, K., 2007. The role of fear appeals in improving driver safety: A review of the effectiveness of fear-arousing (threat) appeals in road safety advertising. International Journal of Behavioral and Consultation Therapy 3, 2003-2222.

Li, Y., Itoh, K., 2013. Safety climate in trucking industry and its effects on safety outcomes. Cognition, Technology & Work, 1-12.

Loeb, P.D., Clarke, W.A., 2007. The determinants of truck accidents. Transportation Research Part E: Logistics and Transportation Review 43, 442-452.

Lu, C.-S., Tsai, C.-L., 2010. The effect of safety climate on seafarers' safety behaviors in container shipping. Accident Analysis & Prevention 42, 1999-2006.

Lu, C.-S., Yang, C.-S., 2010. Safety leadership and safety behavior in container terminal operations. Safety Science 48, 123-134.

Lueck, M., Murray, D., 2011. Predicting truck crash involvement: A 2011 update. American Transportation Research Institute.

Mahmood, F., Ul Asar, A., Mahmood, A., 2006. GPS and remote sensing for emergency vehicle navigation and communication, 2006 International Conference on Advances in Space Technologies, ICAST, September 2, 2006 - September 3, 2006. Inst. of Elec. and Elec. Eng. Computer Society, Islamabad, Pakistan, pp. 33-36.

Makin, A.-M., 2009. Applying the "safe place, safe person, safe systems" framework to improve OHS management: A new integrated approach, School of Risk and Safety Sciences. The University of New South Wales, Sydney, p. 989.

Malterud, K., 2001. Qualitative research: Standards, challenges, and guidelines. The Lancet 358, 483-488.

Manders, S., 2006. "Twice the task" a review of Australia's freight transport tasks. National Transport Commission.

Mayhew, C., Quinlan, M., 2001. Occupational violence in long distance road transport: A study of 300 Australian truck drivers. Current Issues Crim. Just. 13, 36.

Mayhew, C., Quinlan, M., 2006. Economic pressure, multi-tiered subcontracting and occupational health and safety in Australian long-haul trucking. Employee Relations 28, 212-229.

Mayhew, D., Simpson, H., Beirness, D., 2004. Heavy trucks and road crashes. Traffic Injury Research Foundation.

190

McCartt, A., Rohrbaugh, J., Hammer, M., Fuller, S., 2000. Factors associated with falling asleep at the wheel among long-distance truck drivers. Accident Analysis & Prevention 32, 493-504.

McDonald, N., Corrigan, S., Daly, C., Cromie, S., 2000. Safety management systems and safety culture in aircraft maintenance organisations. Safety Science 34, 151-176.

McIntyre, K., Moore, B., 2002. National roa transport compliance and enforcement reforms: On the road to a new national culture of compliance, Current issues in regulation: Enforcement and compliance conference. Australian Institute of Criminology, Melbourne.

McKay, R., 2011. Operation austrans underway nationwide, ATN.

Mearns, K., Whitaker, S., Flin, R., 2003. Safety climate, safety management practice and safety performance in offshore environments. Safety Science 41, 641-680.

Min, H., Lambert, T., 2002. Truck driver shortage revisited. Transportation Journal 42, 5-16.

Mitchison, N., Papadakis, G., 1999. Safety management systems under Seveso II: Implementation and assessment. Journal of Loss Prevention in the Process Industries 12, 43-51.

Monaco, K., Williams, E., 2000. Assessing the determinants of safety in the trucking industry. Journal of Transportation and Statistics 3, 69-80.

Mooren, L., Grzebieta, R., 2011. Review of Australian alternative compliance schemes, TRB Annual Meetings. Transportation Research Board,, Washington DC.

Mooren, L., Grzebieta, R., Williamson, A., 2009a. Lessons from occupational safety for work related road safety, Australasian Road Safety Research, Education and Policing Conference, Sydney.

Mooren, L., Grzebieta, R., Williamson, A., 2009b. Lessons from occupational safety for work related road safety, Australasian Road Safety Research, Policing and Education Conference, Sydney.

Mooren, L., Grzebieta, R., Williamson, A., Olivier, J., 2012. Comparing heavy vehicle safety management in Australia and the United States, ACRS National Conference - a safe system: expanding the reach, Sydney.

Mooren, L., Grzebieta, R., Williamson, A., Olivier, J., Friswell, R., 2014. Safety management for heavy vehicle transport: A review of the literature. Safety Science 62, 79-89.

Mooren, L., Williamson, A., Grzebieta, R., 2015. Evidence that truck driver remuneration is linked to safety outcomes: A review of the literature, Australasian Road Safety Conference 2015. Australasian College of Road Safety, Gold Coast, Australia.

Moreno, C., Carvalho, F., Lorenzi, C., Matuzaki, L., Prezotti, S., Bighetti, P., Louzada, F., Lorenzi-Filho, G., 2004. High risk for obstructive sleep apnea in truck drivers estimated by the berlin questionnaire: Prevalence and associated factors. Chronobiology International 21, 871-879.

Morgan, R., Ogden, K., Barnes, J., 1999. Treatment of crash locations. AustRoads, Sydney.

Morrow, S., McGonagle, A., Dove-Steinkamp, M., Walker, C., Marmet, M., Barnes-Farrell, J., 2009. Relationships between psychological safety climate facets and safety behavior in the rail industry: A dominance analysis. Accident Analysis & Prevention 42, 1460-1467.

Moses, L., Savage, I., 1994. The effect of firm characteristics on truck accidents. Accident Analysis & Prevention 26, 173-179. 191

Mullen, P., Ramirez, G., 2006. The promise and pitfalls of systematic reviews. Annual Review of Public Health 27, 81-102.

Mulrow, C., 1994. Systematic reviews - rationale for systematic reviews. British Medical Journal 309, 597- 599.

Muresan, M., 2007. Fleet monitoring using GPS and DGPS technology. Metalurgia International 12, 27-33.

Murray, W., Ison, S., Gallemore, P., Nijar, H., 2009a. Effective occupational road safety programs: A case study of wolseley, 88th Annual Transportation Research Board Meeting.

Murray, W., Pratt, S., Hingston, J., Dubens, E., 2009b. Promoting global initiatives for occupational road safety: Review of occupational road safety worldwide Department of Health and Human Services CDC & Prevention National Institute for Occupational Safety and Health (NIOSH).

Murray, W., White, J., Ison, S., 2012. Work-related road safety: A case study of roche Australia. Safety Science 50, 129-137.

Murray, W.P., S. and Hingston, J. Dubens, E., 2009. Promoting global initiatives for occupational road safety: Review of occupational safety worldwide, In: NIOSH (Ed.), First International Conference on Road Safety at Work. NIOSH, Washington DC.

Nader, R., 1972. Unsafe at any speed: The designed-in dangers of the American automobile. Grossman Publishers Incorporated, New York.

Naveh, E., Marcus, A., 2007. Financial performance, ISO 9000 standard and safe driving practices effects on accident rate in the U.S. Motor carrier industry. Accident Analysis & Prevention 39, 731-742.

Newnam, S., Goode, N., 2015. Do not blame the driver: A systems analysis of the causes of road freight crashes. Accident Analysis & Prevention 76, 141-151.

Newnam, S., Griffin, M., Mason, C., 2008. Safety in work vehicles: A multilevel study linking safety values and individual predictors to work-related driving crashes. Journal of Applied Psychology 93, 632-644.

Newnam, S., Watson, B., 2011. Work-related driving safety in light vehicle fleets: A review of past research and the development of an intervention framework. Safety Science 49, 369-381.

Nielsen, K., Kines, P., Pedersen, L., Andersen, L., Andersen, D., 2013. A multi-case study of the implementation of an integrated approach to safety in small enterprises. Safety Science.

Nilsen, P., Hudson, D., Kullberg, A., Timpka, T., Ekman, R., Lindqvist, K., 2004. Making sense of safety. Injury Prevention 10, 71-73.

NTARC, 2015. 2015 major accident investigation report. National Truck Accident Research Centre, Brisbane.

NTC, 2004. Information bulletin: Road transport reform (compliance and enforcement) bill, In: National_Transport_Commission (Ed.), Melbourne.

NTC, 2009. Accreditation policy review. National_Transport_Commission, Melbourne.

NTC, 2015. Primary duties for chain of responsibility parties and executive officer liability - discussion paper. National Transport Commission.

O'Neill, S., 2014. Pushing it up the chain: Why big business can't ignore truck safety. The Conversation.

192

OECD, 2008. Towards zero: Ambitious road safety targets and the safe system approach. Organisation for Economic Co-operation and Development (OECD), Paris, France.

Olivier, J., Bell, M., 2013. Effect sizes for 2x2 contingency tables. PLoSONE 8.

Olsen, E., 2010. Exploring the possibility of a common structural model measuring associations between safety climate factors and safety behaviour in health care and the petroleum sectors. Accident Analysis & Prevention In Press, Corrected Proof.

ONRSR, 2013. Guideline for preparation of a rail safety management system. Office of the National Rail Safety Regulator, Adelaide.

Organisation, I.L., 2001. ILO guidelines on occupational safety and health management systems (OHS- ms), In: Organisation, I.L. (Ed.), Geneva.

Öz, B., Özkan, T., Lajunen, T., 2013. An investigation of professional drivers: Organizational safety climate, driver behaviours and performance. Transportation Research Part F: Traffic Psychology and Behaviour 16, 81-91.

Parker, D., West, R., Stradling, S., Manstead, A., 1995. Behavioural characteristics and involvement in different types of traffic accident. Accident Analysis & Prevention 27, 571-581.

Peden, M., Scurfield, R., Sleet, D., Mohan, D., Hyder, A., Jarawan, E., Mathers, C., 2004. World report on road traffic injury prevention, Geneva.

Peignier, I., Leroux, M.-H., de Marcellis-Warin, N., Trapanier, M., 2011. Organizational safety practices of hazardous materials carriers. The International Journal of Transportation Research 3, 149-159.

Perrow, C., 1984. Normal accidents. Basic Books Inc, New York.

Pilkington, P., Kinra, S., 2005. Effectiveness of speed cameras in preventing road traffic collisions and related casualties: Systematic review. BMJ 330, 331-334.

Price Waterhouse Coopers, 2013. A future strategy for road supply and charging in Australia. Australian Trucking Association,.

Quinlan, M., 2001. Report of inquiry into safety in the long haul trucking industry. Motor Accidents Authority, Sydney, p. 353.

Quinlan, M., Bohle, P., 2004. Managing occupational health and safety: A multi-disciplinary approach. Macmillan Publishers, South Yarra.

Quinlan, M., Bohle, P., Lamm, F., 2010. Managing occupational health and safety: A multidisciplinary approach, 3rd ed. Palgrave MacMillan, Melbourne.

Quinlan, M., Mayhew, C., 2001. Evidence versus ideology: Lifting the blindfold on OHS in precarious employment, Working Paper Series.

Quinlan, M., Mayhew, C., Bohle, P., 2001. The global expansion of precarious employment, work disorganization, and consequences for occupational health: A review of recent research. International Journal of Health Services 31, 335-414.

Quinlan, M., Wright, L., 2008a. Remuneration and safety in the Australian heavy vehicle industry: A review undertaken for the National Transport Commission, Melbourne.

193

Quinlan, M., Wright, L., 2008b. Safe payments: Addressing the underlying causes of unsafe practices in the road transport industry. National Transport Commission, Melbourne.

Rasmussen, J., 1997. Risk management in a dynamic society: A modelling problem. Safety Science 27, 183-213.

Rawling, M., Kaine, S., 2012. Regulating supply chains to provide a safe rate for road transport workers. Australasian Journal of Labour Law 25, 237-257.

Reason, J., 1990. Human error. Cambridge University Press.

Reason, J., 1997a. Managing the risks of organisational accidents. Ashgate Publishing, Aldershot.

Reason, J., 1997b. Presentation to the royal military college, RMC-V, Cleveland.

Reason, J., 2000. Human error: Models and management. British Medical Journal 320, 768-770.

Reason, J., Shotton, R., Wagenaar, W., Hudson, P., Groeneweg, J., 1989. Tripod, a principled basis for safer operations. Shell Internationale Petroleum Maatschappij, The Hague.

Rechnitzer, G., Grzebieta, R., 1999. Crashworthy systems - a paradigm shift in road safety design. IEAust 5.

Richards, N., 2004. Fatigue and beyond: Patterns of, and motivations for illicit drug use among long haul truck drivers, Centre for Accident Research and Road Safety-Queensland (CARRS-Q), School of Psychology and Counselling. Queensland University of Technology, Brisbane.

Roberts, K., 1990. Some characteristics of one type of high reliability organization. Organization Science 1, 160-176.

Robotham, G., 2001. Safety training that works. Professional Safety 46, 33-37.

Robson, L., Clarke, J., Cullen, K., Bielecky, A., Severin, C., Bigelow, P., Irvin, E., Culyer, A., Mahood, Q., 2007. The effectiveness of occupational health and safety management system interventions: A systematic review. Safety Science 45, 329-353.

Rodríguez, D., Rocha, M., Khattak, A., Belzer, M., 2003. Effects of truck driver wages and working conditions on highway safety: Case study. Transportation Research Record: Journal of the Transportation Research Board 1833, 95-102.

Rodriguez, D., Targa, F., Belzer, M., 2006. Pay incentives and truck driver safety: A case study. Industrial and Labor Relations Review 59, 205-225.

Roland, H., Moriaty, B., 1990. System safety, engineering and management, Second ed. John Wiley and Sons, Hoboken, New Jersy.

Rufford, P., Baas, P., 2006. Policy review of road transport heavy vehicle accreditation: Discussion paper, In: Commission, N.T. (Ed.). National Transport Commission, Melbourne.

Runyan, C., 1998. Using the Haddon matrix: Introducing the third dimension. Injury Prevention 4, 302-307.

Sabbagh-Ehrlich, S., Friedman, L., Richter, E.D., 2005. Working conditions and fatigue in professional truck drivers at israeli ports. Injury Prevention 11, 110-114.

Sabey, B., Taylor, H., 1980. The known risks we run: The highway, In: Schwing, R., Albers, W., Jr. (Eds.), Societal risk assessment. Springer US, pp. 43-70.

194

Safe Work Australia, 2013. Work health and safety in the road freight transport industry.

Safe Work Australia, 2014a. Work-related fatalities involving trucks, Australia, 2003 to 2012, Canberra.

Safe Work Australia, 2014b. Work-related traumatic injury fatalities, Australia 2013, Canberra.

Safe Work Australia, 2014c. Work-related traumatic injury fatalities, Australia 2013, Canberra.

Saksvik, P., Torvatn, H., Nytro, K., 2003. Systematic occupational health and safety work in Norway: A decade of implementation. Safety Science 41, 721-738.

Salminen, S., 2008. Two interventions for the prevention of work-related road accidents. Safety Science 46, 545-550.

Salmon, P., Lenné, M., 2015. Miles away or just around the corner? Systems thinking in road safety research and practice. Accident Analysis & Prevention 74, 243-249.

Salmon, P.M., McClure, R., Stanton, N.A., 2012. Road transport in drift? Applying contemporary systems thinking to road safety. Safety Science 50, 1829-1838.

Santos-Reyes, J., Beard, A., 2002. Assessing safety management systems. Journal of Loss Prevention in the Process Industries 15, 77-95.

Senserrick, T., Whelan, M., 2003. Graduated driver licensing: Effectiveness of systems and individual components. MUARC.

Seo, D., Torabi, M., Blair, E., Ellis, N., 2004. A cross-validation of safety climate scale using confirmatory factor analytic approach. Journal of Safety Research 35, 427-445.

Sheridan, T., 2008. Risk, human error, and system resilience: Fundamental ideas. Human Factors: The Journal of the Human Factors and Ergonomics Society 50, 418-426.

Short, J., Boyle, L., Shackelford, S., Inderbitzen, B., Bergoffen, G., 2007. The role of safety culture in preventing commercial motor vehicle crashes: A synthesis of practice, In: Board, T.R. (Ed.), Commercial Truck and Bus Safety Synthesis Program. Transport Research Board, Washington DC.

Silva, S., Lima, M., 2005. Safety as an organisational value: Improving safety practices, In: Kolowrocki. (Ed.), Advances in safety and reliability. Taylor & Francis Group, London, pp. 1817-1824.

Silva, S., Lima, M., Baptista, C., 2004. OSCI: An organisational and safety climate inventory. Safety Science 42, 205-220.

Skinner, S., 2015. Victoria Police to beef up CoR skills, Australian Transport Network,.

Sklet, S., 2004. Comparison of some selected methods for accident investigation. Journal of Hazardous Materials 111.

Stuckey, R., LaMontagne, A., Sim, M., 2007. Working in light vehicles--a review and conceptual model for occupational health and safety. Accident Analysis & Prevention 39, 1006-1014.

Svanstrom, L., 2000. Evidence-based injury prevention and safety promotion: State-of-the-art, In: Mohan, D., Geetam, T. (Ed.), Injury prevention and control. Taylor & Francis, pp. 181-198.

Swedish Ministry of Transport and Communications, 1997. En route to a society with safe road traffic.

195

Thompson, J., Stevenson, M., 2014. Associations between heavy vehicle driver compensation methods, fatigue-related driving behavior, and sleepiness. Traffic Injury Prevention 15, 10-14.

Tingvall, C., 1998. The Swedish "Vision Zero" and how parliamentary approval was obtained, Australasian Road Safety Research, Policing, Education Conference, Wellington, New Zealand.

Tingvall, C., Haworth, N., 1999. Vision Zero - an ethical approach to safety and mobility, 6th ITE International Conference Road Safety & Traffic Enforcement: Beyond 2000, Melbourne.

Transport Workers Union of Australia, 2012. Make our roads safer for all australians.

U.S. Department of Transportation, 2015. 2015 pocket guide to large truck and bus statististics, Washington DC. van Schagen, I., Janssen, T., 2000. Managing road transport risks: Sustainable safety in the Netherlands. IATSS Research 24, 18-27.

Verbeek, J., Ruotsalainen, J., Hoving, J., 2012. Synthesizing study results in a systematic review. Scandinavian Journal of Work and Environmental Health 38, 282-290.

Verlinden, J., 2002. Sqas: Safety and quality assessment systems for the transport/storage/handling of chemicals. THE HANDBOOK OF HAZARDOUS MATERIALS SPILLS TECHNOLOGY 1.

Vredenburgh, A., 2002. Organizational safety: Which management practices are most effective in reducing employee injury rates? Journal of Safety Research 33, 259-276.

Wachter, J., Yorio, P., 2014a. A system of safety management practices and worker engagement for reducing and preventing accidents: An empirical and theoretical investigation. Accident Analysis & Prevention.

Wachter, J.K., Yorio, P.L., 2014b. A system of safety management practices and worker engagement for reducing and preventing accidents: An empirical and theoretical investigation. Accident Analysis & Prevention 68, 117-130.

Wåhlberg, A., Dorn, L., Kline, T., 2010. The effect of social desirability on self reported and recorded road traffic accidents. Transportation Research Part F: Traffic Psychology and Behaviour 13, 106-114.

Walters, D., James, P., 2011. What motivates employers to establish preventive management arrangements within supply chains? Safety Science 49, 988-994.

Wang, L., Pei, Y., 2014. The impact of continuous driving time and rest time on commercial drivers' driving performance and recovery. Journal of Safety Research 50, 11-15.

Ward, J., Warren, C., (eds.), 2006. Silent victories: The history and practice of public health in twentieth century america. Oxford University Press.

Wegman, F., Atze, D., Schermers, G., van Vliet, P., 2005. Sustainable safety in the Netherlands: The vision, the implementation and the safety effects, 3rd International Symposium on Highway Geometric Design. SWOV Chicago, Illinois.

Weick, K.E., Sutcliffe, K.M., Obstfeld, D., 2008. Organizing for high reliability: Processes of collective mindfulness, In: Boin, A. (Ed.), Crisis management. Sage, London, pp. 81-123.

WHO, 2010. Global plan for the Decade of Action for road safety 2011-2020. World Health Organisation, Geneva.

196

WHO, 2015. Global status report on road safety 2015, Geneva.

Williamson, A., 2005. Fatigue and coping with driver distraction. Journal of the Australasian College of Road Safety, 19-21.

Williamson, A., 2007. Predictors of psychostimulant use by long-distance truck drivers. American journal of epidemiology 166, 1320-1326.

Williamson, A., Bohle, P., Quinlan, M., Kennedy, D., 2009. Short trips and long days: Safety and health in short-haul trucking. Industrial & Labor Relations Review 62.

Williamson, A., Feyer, A., 2000. Moderate sleep deprivation produces impairments in cognitive and motor performance equivalent to legally prescribed levels of alcohol intoxication. Occupational and Environmental Medicine 57, 649-655.

Williamson, A., Feyer, A., Cairns, D., Biancotti, D., 1997. The development of a measure of safety climate: The role of safety perceptions and attitudes. Safety Science 25, 15-27.

Williamson, A., Feyer, A., Friswell, R., Saduri, S., 2001. Driver fatigue: A survey of professional heavy vehicle drivers in Australia. National Road Transport Commission, Melbourne.

Williamson, A., Friswell, R., 2013. The effect of external non-driving factors, payment type and waiting and queuing on fatigue in long distance trucking. Accident Analysis & Prevention 58, 26-34.

Williamson, A., Irvine, E., Friswell, R., 2003. What is the involvement of heavy trucks in crashes in NSW?, Road Safety Research, Policing and Education Conference, Sydney.

Wills, A., Biggs, H., Watson, B., 2005. Analysis of a safety climate measure for occupational vehicle drivers and implications for safer workplaces. Australian Journal of Rehabilitation Counselling 11, 8-21.

Wolf, F., 2001. Operationalizing and testing normal accident theory in petrochemical plants and refineries. Production and Operations Management 10, 292-305.

WorkCover NSW, 2015. Safety in the road freight transport industry, Sydney.

Wouters, P., Bos, J., 2000. Traffic accident reduction by monitoring driver behaviour with in-car data recorders. Accident Analysis & Prevention 32, 643-650.

Wright, J., Walton, J., 2006. Transport industry- mutual responsibility for road safety (state) award and contract determination, In: Industrial Relations Commission of NSW (Ed.), Sydney.

Wright, J., Walton, J., Sams, D., Tabbaa, C., 2006. Transport industry - mutual responsibility for road safety. Industrial Relations Commission of New South Wales,, Sydney.

Zohar, D., 1980. Safety climate in industrial organizations: Theoretical and applied implications. J Appl Psychol 65, 96-102.

Zohar, D., 2002. The effects of leadership dimensions, safety climate, and assigned priorities on minor injuries in work groups. Journal of Organizational Behavior 23, 75-92.

Zohar, D., Huang, Y.-H., Lee, J., Robertson, M., 2014. A mediation model linking dispatcher leadership and work ownership with safety climate as predictors of truck driver safety performance. Accident Analysis & Prevention 62, 17-25.

197

Zohar, D., Polachek, T., 2014. Discourse-based intervention for modifying supervisory communication as leverage for safety climate and performance improvement: A randomized field study. Journal of Applied Psychology 99, 113-124.

Zwetsloot, G., Hale, A., Zwanikken, S., 2011. Regulatory risk control through mandatory occupational safety and health (OSH) certification and testing regimes (ctrs). Safety Science 49, 995-1006.

198

Appendix A – Ten-Point National Logistics Safety Code

199

Appendix B- Manager Survey Questionnaire

Freight and vehicle fleet

1. What type of freight do your drivers carry? Please tick all that apply: D Building materials D Manufactured goods (e.g., clothing) D Car carrying What type of goods? D Containers D Dangerous materials (e.g.,fuel, D Bulk (e.g., grain, coal, quarry materials} chemicals} What type of bulk? D Express freight D Farm produce D Refrigerated or temperature controlled D General/mixed freight D Removals D Groceries D Other - Please specify D Livestock D Machinery

2. How many trucks does your company operate in each of the following categories? Number of trucks (If none, tick box) Superlight (less than 4.5t GVM) D Light (4.5-12t GVM) D Heavy rigid (over 12 t GVM) D Heavy articulated D B-Double D Road Train Type 1 (B-Triple) D Road Train Type 2 D Performance Based Standards (PBS) D

Other (Specify) ~====~-~===~ ______~, D,J_,______j

3. How many insurance claims for truck incidents did your company make between 1 January, 2007 and 31 December, 2009 (excluding claims that were damage from "natural occurrence" or third party at fault}? ______claims

Vehicles acquisition and maintenance

4. Do you usually purchase new vehicles? D Yes D No If your answered No, how old are the vehicles you usually purchase? ______years

5. On average, how long do you keep heavy trucks in your fleet? ______years

6. Does your company have a truck disposal policy? D Yes D No If yes, please advise what it is------

7. Whether purchased new or used, what is the average age of the heavy trucks in your fleet ______years

8. Does your company have a vehicle purchasing policy? D Yes D No

200

If you answered Yes, does it include safety considerations? What safety features are included Anti-lock braking system Adaptive speed control Airbags Air disc brakes Cab strength to ECE 29 Standard Driver access to cabin Electronic braking system (EBS) Electronic stability program (ESP) Front underrun device Integrated seatbelt/suspension seat Lane assist device Load securing devices Noise level, comfort, ventilation Rear underrun device Safety access to and from vehicle Seat adjustment Side underrun device Tyre pressure monitoring systems Blind spot visibility around vehicle devices Other (specify) ______If you answered No, do your heavy truck purchasing decisions include consideration of safety features? What safety features are considered? (please tick features above)

9. Do you have regular scheduled maintenance of all heavy trucks in the fleet? Yes No If you answered Yes, how often are the heavy trucks maintained? Per ______km? OR Per ______weeks Is this schedule: More often than the manufacturer’s recommendation According to the manufacturer’s recommendation Less frequent than the manufacturer’s recommendation

10. What was the total fuel consumption of your heavy truck fleet in 2010? ______litres

11. In total across the fleet, how many days were heavy trucks off road due to mechanical problems in 2010? ______days

Staffing

12. How many heavy vehicle drivers work for your company in the following categories? Employees ______(Sub)contractors in company colours ______Freelance (sub)contractors (used regularly) ______

13. How many management staff work for your company? ______(If transport and storage is only one part of your company’s operations, include only managers in the transport and storage sections of the business)

201

Journey and Risk Assessment

14. Please estimate what percentage of your heavy truck trips are driven by each category of driver: Employees (Sub)contractors Freelance Trips less than 100km ______% ______% ______% Trips between 100-750 km ______% ______% ______% Trips between 750-1,000km ______% ______% ______% Trips greater than 1,000kms ______% ______% ______%

15. Does your route selection and journey planning include safety considerations? Yes No Don’t have a policy on this If Yes, what safety features are included? (Tick all that apply) Grade separation Overtaking lanes Bridge capacity Road conditions (shoulders, potholes, tactile line-marking, etc) Rest area availability HAZMAT routes Over-dimensional vehicle access Speed limiting on poorer quality roads Traffic conditions Weigh stations Safety-Cams Tunnels and low underpasses Other (specify)______

16. Do you conduct risk assessments or safety audits at: Your own depot/s? Yes No Delivery sites? All Most Some Very few None

Driver Recruitment

17. What process do you use to recruit and select heavy truck drivers? Tick all options that apply Recruitment or HR firm In-house (Sub)contractors are responsible for recruiting drivers

18. Do you require heavy truck drivers to have any specific qualifications or certificates in order to work for your company? Yes What do you require? ______No

19. What aspects of a driver’s history and performance are routinely checked prior to their working for your company? (Tick all options that apply) References from previous employers Licence currency Licence demerit points Accident history Selection test performance In-vehicle driving performance Health

202

None of the above. Comment: ______

20. How many employee heavy truck drivers at your company are in the following age categories? Under 25 years of age ______26-55 years of age ______55-65 years of age ______over 65 years of age ______

Pay and Conditions

21. How are your drivers paid for driving? Employees (Sub)contractors Freelance Hourly rate Flat day rate Day rate with overtime Weekly rate Weekly rate with overtime Salary Flat rate per truckload Trip rate (based on kms travelled or tonnage carried) Other (specify) ______

22. How are your drivers paid for the following non-driving work activities? Employees (Sub)contractors Freelance Loading and unloading Not paid Paid for all hours spent Drivers do not load/unload Other (specify) ______

Employees (Sub)contractors Freelance Queuing/waiting to load and unload Not paid Paid for all hours spent Other (specify)

23. Do (sub)contracted heavy truck drivers do the same work as employees drivers at your company? Same work Different work If Different, how do they differ? ______

Policies

24. Do you have policies on any of the following issues: Tick all options that apply OHS risk assessment OHS audits OHS reporting systems Fleet management Driver recruitment and selection

203

Driver training Fitness for duty Fatigue management On-road behaviour Depot behaviour Work monitoring Work planning (e.g, journey planning) Driver performance monitoring Vehicle selection Vehicle maintenance Depot conditions Accident prevention and response If your answered yes to any of the above, could we see the policy documents? Yes (Please send to Lori Mooren, Transport and Road Safety, Old Main Bldg, University of NSW, Kensington, 2052) No

25. Are employee and (sub)contractor truck drivers subject to all of the same safety policies? Yes, all drivers whether employees or (sub)contractors are subject to all of the same policies

No, they differ in that ______No, our policies do not apply to (sub)contractors Do the policies of the (sub)contractor firms apply to the (sub)contractor drivers working at your company? Yes No I don’t know If Yes, how do the (sub)contractor policies differ from your company’s policies for employee drivers? (Sub)contractor policies tend to be stricter than company policies (Sub)contractor policies are similar to company policies (Sub)contractor policies tend to be less strict than company policies I don’t know what is in the policies of firms that have been (sub)contracted

26. Do all drivers sign off on agreed codes of conduct (eg a driver manual, policy statement)? Employees (Sub)contractors Freelance Yes Yes Yes No No No

27. Do managers have heavy vehicle safety management KPIs (Key Performance Indicators) or performance criteria at your company? Yes No Don’t know Is your company accredited or certified under any of the following schemes? Tick all options that apply NHVAS (National Heavy Vehicle Accreditation Scheme) Mass Management Maintenance BFM (Basic Fatigue Management) AFM (Advanced Fatigue Management) TruckSafe ISO 9001 (International Standards Organisation - Quality management systems) Other Which schemes are these? ______

204

Scheduling

28. How are driver schedules and rosters managed? Tick all options that apply Schedules and rosters are determined centrally Local depot schedules and rosters drivers Loads are scheduled centrally and rosters are determined locally Rostering/scheduling software is used (Please specify) ______Other (Please specify) ______

29. Are schedulers trained in fatigue risk management? Yes What type of training do you provide? ______No

30. How are hours and schedules monitored? Hours and schedules are not monitored In-truck logs or work diary records are reviewed after an incident Random spot checks are carried out of in–truck or work diary records All in-truck recordings are reviewed All work diary records are reviewed Other Please describe other ______

Training

31. What kind of safety training is provided for drivers in your company? Employees (Sub)contractors Freelance Occupational Health & Safety Fatigue risk management Driving skills (on road) Driving skills (class room) Eco driving (fuel economy) Manual handling Loading/unloading Other (specify) ______

32. Do any of your staff who are responsible for safety management have specific training in OHS (Occupational Health & Safety) or SHE (Safety, Health & Environment)? Yes What role do they occupy? ______What kind of training have they had? ______No

33. Does your company provide any training for managers on safety issues (e.g., fatigue, mass or load requirements, OHS consultation, hazardous substances etc)? Yes What type of training do you provide? ______

34. How is safety information communicated to drivers working for the company? Tick all options that apply Newsletter

205

Noticeboard Supervisors Union representatives Toolbox talks or staff meetings Special briefings Text messages Email Staff website Other Please describe other ______

Driver Participation in OHS

35. In your company, can drivers be involved in OHS (Occupational Health & Safety) decision-making? Yes No Don’t know If you answered Yes How can drivers participate? (i.e., what formal mechanisms are in place?) Tick all options that apply OHS committee Through an OHS representative Through a union representative Toolbox or staff meetings Suggestion box Other What are these other mechanisms? ______

36. What is the process for dealing with safety concerns raised by drivers? ______Is there a time limit for dealing with drivers’ safety concerns? Yes No Is this limit monitored and enforced? Yes No Is it mandatory to provide feedback to drivers about action taken? Yes No

Incidents and Record Keeping

37. How many injuries were recorded for heavy truck drivers at your company in last year? Number of lost time injuries ______Number of non-lost time injuries ______

38. What was the lost time injury rate per total hours worked for heavy truck drivers in last year? ______

39. Does your company keep records of on-road violations for individual drivers and vehicles? Tick all that are kept for: Employees (Sub)contractors Freelance Driver infringements Vehicle defect notices

206

Driver management and discipline

40. Does your company use any form of in-vehicle driver monitoring Tick all options that apply GPS Fuel consumption Braking analysis Gear change analysis Speed analysis Fatigue monitoring systems (e.g., Optalert) Other Please describe other ______41. What is the company procedure for dealing with breaches of working hours? Nothing. It is the driver’s responsibility Warn the driver to avoid this in future Formally discipline & penalise the driver Dismiss the driver on the spot Investigate the reasons for the breach Other Please describe other ______42. What action is usually taken when people behave unsafely? Tick all options that apply Formal investigation Disciplinary procedures Education or training Other Please describe other actions ______

43. Do you have a system for analysing incidents, accidents and crashes? Yes What is this system? ______No

44. Does your company provide positive incentives for safe work or management practices? Yes What incentive systems do you use? ______

CONTACT NAME of Organisation’s representative ______

207

Appendix C - Recording sheet for in-depth investigation

Company No.____

Questions Evidence describe what was confirmed Implemented/description describe the by: observations (o), documents (d), manifestations or implementation of policies or verbal response by manager (vm), practices verbal response by driver (vd)

1a. Have there been business or organisational changes to: Freight carried 1b. Have there been business or organisational changes to: Fleet vehicles

1c. Have there been business or organisational changes to: Managers/management systems 1d. Have there been business or organisational changes to: Driver/ staff 1e. Have there been business or organisational changes to: Pay/conditions

1f. Have there been business or organisational changes to: Journey & site risks 1g. In the last 12 months number of: Insurance claims

1h. In the last 12 months number of: Lost time injuries

1i. In the last 12 months number of: No-lost time injuries

1j. In the last 12 months number of: Defect notices

208

1k. In the last 12 months number of: Infringements

1l. In the last 12 months number of: Days off the road due to mechanical breakdown

2a. Does this company make any special safety provisions for delivery of the type of freight carried?

2b. Does the company require specific safety features in its fleet purchasing practices?

2c. How is it decided when to dispose of a truck?

2d) How often does this company send trucks in for a routine maintenance service?

2e. When a driver initiates a truck repair, what is the response or procedure?

3a. What is the recruitment process for drivers & what pre- employment checks are undertaken? 3b. Does this company employ drivers under the age of 25 and/or over the age of 60?

3c. What are the driving remuneration methods?

209

3d. What are the loading/unloading remunerations methods?

3d. What are the queuing/waiting remunerations methods?

4a. What does the company do to assess the driving risks on delivery journeys?

4b. How and when are site risk assessments conducted?

4c. How is fatigue risk taken into account when scheduling journeys or assigning work rosters to drivers? 4d. If fatigue risk training is provided, who receives the training?

4e. How is OHS data used by this company?

4f. How are drivers informed about this company’s OHS policies?

4g. How can drivers participate in OHS decision-making?

4h. How does this company ensure that drivers are fit for duty?

4i. How does this company assure safe driving behaviour?

4j. How does this company assure safe freight handling?

210

4k. How is driver behaviour monitored by this company?

4l. How are drivers disciplined for not complying with safety policies?

4m. What is the procedure for responding to a crash or adverse incident?

4n. How is safety management performance assessed?

4o. How are staff acknowledged or provided incentives for safety innovations?

5a. How comprehensive is the safety management system of this company?

211

Appendix D - Driver Survey Questionnaire

Company:______Name:______Phone no.______

15. What type of freight do you carry? Please tick all that apply: Building materials Manufactured goods (e.g., clothing) Car carrying What type of goods? Containers ______Dangerous materials (e.g.,fuel, Bulk (e.g., grain, coal, quarry chemicals) materials) What type of bulk? Express freight ______Farm produce Refrigerated or temperature controlled General/mixed freight Removals Groceries Other – Please specify Livestock ______Machinery

16. What type of truck do you usually drive? Superlight (less than 4.5t GVM) Light (4.5-12t GVM) Heavy rigid (over 12 t GVM) Heavy articulated B-Double Road Train Type 1 (B-Triple) Road Train Type 2 Performance Based Standards (PBS) Other (Specify) ______

17. Does the truck you usually drive have any of the features listed below? : Y=yes, N=no (tick one of the three choices for each feature listed in appropriate box for yes, no, or don’t know) Y N Don’t know Y N Don’t know ABS Integrated seatbelt/suspension seat Adaptive Speed Control Lane Assist device Airbag Load securing devices Air conditioning Noise level, comfort, ventilation Air disc brakes Rear Under-run Device Automatic collision notification Safety access to and from the vehicle system Seat adjustment Cab strength to ECE 29 standard Side Under-run Device Collapsible steering wheel Tyre pressure monitoring systems Electronic Braking System Blind spot visibility around vehicle Electronic Stability Program devices Front Under-run Device Other – ______18. How old is the truck you usually drive for the company? ______years 19. How well maintained is the truck you normally drive? Excellent Very well Satisfactory Less than satisfactory Poorly 20. On average how often does the truck you normally drive break down? Less than once per year Once per year Once per 6 months Once per month At least once per fortnight

21. How far are your normal trips each day? Less than 100km 100-1000km Greater than 1000km

212

22. What is your age? ______years old 23. How are you paid for driving? (tick appropriate box) Hourly rate Flat day rate Day rate with overtime Weekly rate Weekly rate with overtime Salary Flat rate per truckload Trip rate (based on kms travelled or tonnage carried) Other (specify) ______

24. How are you paid for the following non-driving work activities? Loading and unloading Not paid Paid for all hours spent I do not load/unload Other (specify) ______

Queuing/waiting to load and unload Not paid Paid for all hours spent I do not queue/wait Other (specify) ______

25. If there are (sub)contracted heavy truck drivers and employee drivers, do the (sub)contracted drivers do same work as employee drivers at your company? Same work Different work Not applicable If Different, how do they differ? ______

213

26. Does this company have policies on any of the following issues? Have you seen them? Have you been asked to sign off that you have read them? Tick a box in each 3 parts of the question.

Safety Policies The company I have seen it I signed off on it has this Be sure to complete all 3 parts Yes No ? Yes No ? Yes No ? OHS risk assessment policy OHS audits policy OHS reporting policy Fleet management policy Driver selection policy Driver training policy Fitness for duty policy Fatigue management policy On-road behaviour policy Depot behaviour policy Work monitoring policy Seatbelt policy Work planning (e.g, trip planning) policy Driver performance monitoring policy Vehicle selection policy Vehicle maintenance policy Depot conditions policy Accident prevention and response policy Comments: ______27. Is this company accredited or certified under any of the following schemes? Tick all options that apply Y N Don’t know NHVAS Mass Management NHVAS Maintenance NHVAS BFM (Basic Fatigue Management) NHVAS AFM (Advanced Fatigue Management) TruckSafe ISO 9001 (International Standards Organisation - Quality management systems) Other Which schemes are these? ______28. How are your schedules and rosters managed? Tick all options that apply Schedules and rosters are determined centrally Local depot schedules and rosters drivers Loads are scheduled centrally and rosters are determined locally Rostering/scheduling software is used (Please specify)______Other (Please specify) ______

29. How are your hours and schedules monitored? Tick all options that apply

214

Hours and schedules are not monitored In-truck logs or work diary records are reviewed after an incident Random spot checks are carried out of in–truck or work diary records All in-truck recordings are reviewed All work diary records are reviewed Other Please describe other

30. What kind of safety training is provided for drivers in your company? What is your experience of the training at the company in the past 12 months?

Safety Training Company Conducted in Attended Improved my provides this past 12 in past 12 safety in past Be sure to complete all 4 parts training months months 12 months Yes No ? Yes No ? Yes No Yes No ? Occupational Health & Safety training Fatigue risk management training Driving skills (on road) training Driving skills (class room) training Eco driving (fuel economy) training Manual handling training Loading/unloading training On-road behaviour training Other safety training (specify):

______Comments: ______

31. How is safety information communicated to you from the company? What is your experience of the safety communications in the past 12 months? Safety Information Safety Received in Useful in Improved my Communicated information past 12 past 12 safety in past is provided months months 12 months Be sure to complete all 4 parts via.. Yes No ? Yes No Yes No ? Yes No ? Newsletter Notice board Supervisors Union representatives Toolbox talks/meetings Special briefings Text messages Email Staff website Other communications (specify): ______Comments: ______

215

32. In this company, how can drivers be involved in OHS (Occupational Health & Safety) decision-making? What is your experience of OHS involvement in the past 12 months? Driver involvement in OHS Involvement Occurred in Acted on by Improved my decision-making, through: is possible past 12 managemen safety in past via.. months t in past 12 12 months Be sure to complete all 4 parts months Yes No ? Yes No ? Yes No ? Yes No ? OHS Committee An OHS representative A union representative Toolbox talks/meetings Suggestion box Text messages Email Direct to management Other method(s) (specify): ______Comments: ______

33. What is the process for dealing with safety concerns raised by drivers? ______Is there a time limit for dealing with drivers’ safety concerns? Yes No Don’t know Is this limit monitored and enforced? Yes No Don’t know Is it mandatory to provide feedback to drivers about action taken? Yes No Don’t know

34. Is the truck you normally drive fitted with any form of in-vehicle driver monitoring? Tick all options that apply My truck has: It is It’s useful Improves my Be sure to complete all 3 parts operating Yes No ? safety Yes No ? Yes No ? GPS tracking A fuel consumption monitor A braking analysis monitor A gear change analysis monitor A speed analysis monitor A fatigue monitoring device Other (specify) ______

Comments: ______

35. What is the company procedure for dealing with breaches of working hours? For breaches of working hours The company It’s Improves my

216

limits does effective safety Yes No ? Yes No ? Yes No ? Be sure to complete all 3 parts Nothing. It is the driver’s responsibility Warn the driver to avoid this in future Formally discipline & penalise the driver Dismiss the driver on the spot Investigate the reasons for the breach Other ______

Comments: ______

217

36. What action is usually taken when people behave unsafely? Tick all options that apply When people behave unsafely The company It’s Improves my does effective safety Be sure to complete all 3 parts Yes No ? Yes No ? Yes No ? Nothing. It is the driver’s responsibility Warn the driver to avoid this in future Formally investigation Disciplinary procedures Education or training Dismiss the driver on the spot Other ______Comments: ______

37. Are all safety incidents investigated in your company? Yes No Don’t know If yes, how? ______

38. Does your company provide positive incentives for safe work practices? Yes No Don’t know If yes, what are the incentives? ______

39. Is there anything else you would like to say about safety management at this company? Comments: ______

Please contact Lori Mooren on 0412 888 290 if you wish to discuss anything about this survey.

Thank you!!!

218