Productivity in Construction Dozzi, S.P.; AbouRizk, S.M.

NRCC-37001

The material in this document is covered by the provisions of the Copyright Act, by Canadian laws, policies, regulations and international agreements. Such provisions serve to identify the information source and, in specific instances, to prohibit reproduction of materials without written permission. For more information visit http://laws.justice.gc.ca/en/showtdm/cs/C-42

Les renseignements dans ce document sont protégés par la Loi sur le droit d’auteur, par les lois, les politiques et les règlements du Canada et des accords internationaux. Ces dispositions permettent d’identifier la source de l’information et, dans certains cas, d’interdire la copie de documents sans permission écrite. Pour obtenir de plus amples renseignements : http://lois.justice.gc.ca/fr/showtdm/cs/C-42

Productivity in Construction

S.P. Dozzi, P.Eng. and S.M. AbouRizk, Ph.D., P.Eng. Construction Engineering and Management Civil Engineering Department University of Alberta

Institute for Research in Construction National Research Council Ottawa, Ontario, Canada IRC-P-3547 NRCC 37001 NR16-2411993E ISBN 0-662-21134-0 Ottawa, December 1993 ©National Research Council Canada 1993 Preface

NRC' Institute for Research in Con­ struction i · delighted to be instrumental in bringing thi important gujdebook to the con- truction profe ional of Canada. Thi CSCE­ NRC project i a good example of lhe kind o.f alliance that i increa ingly important in up­ port of lhe Canadian construction indu try as it strives to find internationally competitive ways to do bu. ine . The pairing of the counlry' leading ource for construction technol.ogie with the enior national profe ional a ocia­ tion in civil engineering ha produced a defini­ tive . tatement on the ubject of productivity, which I commend to all reader .

G. Seaden, Director General Jn ·titote for Research in Construction

PRODUCTIVITY IN CONSTRUCTION

Foreword

Contractors have often been heard to In response to this dilemma, the Con­ say, "As long as we are as aggressive and effi­ struction Division of the Canadian Society for cient as our usual competitors, we will always Civil Engineering (CSCE) developed and im­ get our share of work." But in today's market­ plemented a program with a view to improving place, being as efficient as one's neighbour productivity. CSCE, with the assistance of the does not suffice. Competition is no longer lim­ National Research Council, formed an alliance ited to contractors working in well defined geo­ with the Construction Technology Centre At­ graphical areas. The available work is being lantic Inc. (CTCA), according to which CSCE sought by firms from other parts of the country would produce a manual about ways to im­ or even of the globe. prove productivity and CTCA would organize seminars. Such seminars on productivity im­ Canadian competitiveness, or rather provement have taken place across Canada the lack of it, has been in the headlines now for since September 1990, usually in collaboration several years. For example, a report in the 25 with the local construction association. June 1991 issue of The Economist, entitled "A Survey of Canada," claims that: The Institute for Research in Con­ struction has now decided to draw on the expe­ "In general, the growth of Canadian produc­ rience gained from the preparation of the tivity is declining; yet if Canada is to remain manual and presentations, and publish this doc­ a high-wage economy, it has to be a high­ ument, "Productivity in Construction." I hope productivity one. Annual productivity that it receives the attention it deserves and that growth, which has been 2.3% in 1946-73,fell every supervisor of construction projects refers to 0.9% in 1973-90. And the growth of Cana­ to it frequently for guidance. dian manufacturing productivity has slowed relative to all other members of the Group of Seven rich countries. Cost competitiveness relative to the United States has declined particularly sharply .. .. " There are also signs of slowed produc­ Stephen G. Revay, F.EIC, F.CSCE tivity in Canada relative to Japan. Between Past President ( 1989-90) CSCE 1986 and 1990, the productivity of construction labour in Japan increased by 6.6% a year, while Canadian construction productivity rose by only 1.6% .

• PRODUCTIVITY IN CONSTRUCTION

~ Table of Contents

1 Introduction 3.3.4 Crews and teamwork ...... 23 3.3.5 Environmental factors ...... 23 1.1 Productivity: More Achievement 3.3.6 Workspace ...... 24 per Resource ...... 1 3.4 Job Site Planning ...... 24 1.2 What is Productivity? ...... 1 3.4.1 Job site planning considerations ...... 24 1.3 Framework for Productivity 3.4.2 Temporary electrical service ...... 24 Improvement in Construction ...... 2 3.4.3 Temporary heating and hoarding ...... 25 1.4 Organization of this Publication ...... 2 3.4.4 Miscellaneous systems ...... 25 3.4.6 Offices, lunchrooms, and sanitary 2 Techniques for Measur­ facilities ...... 25 ing and Improving 3.5 Safety Issues ...... 25 Productivity at 3.5.1 Economic impact of accidents ...... 25 Construction Sites 3.5.2 Safety and productivity ...... 25 2.1 Introduction ...... 5 4 Measuring Productivity 2.2 Measuring and Interpreting from the Cost-Reporting Work and Crew Effectiveness ...... 5 2.2.1 Field rating ...... 5 System 2.2.2 W ark sampling ...... 5 4.1 Introduction ...... 27 2.2.3 Five-minute rating ...... 7 4.2 Data Collection and Processing ...... 27 2.3 Field Surveys ...... 7 4.3 Tracking Person-hours instead of 2.3.1 Foreman delay survey ...... 8 Costs in the Cost-reporting System ... 30 2.3.2 Craftsman questionnaire ...... 8 4.3.1 Estimated 'percent complete' ...... 30 2.4 The Method Productivity Delay 4.3.2 Physical measurement ...... 31 Model ...... 9 4.3.3 Earned value ...... 31 2.5 Charting Techniques: 4.3.4 Performance factors ...... 31 Crew-Balance Charts ...... 12 4.4 Cost Reporting and Analysis Using 2.6 Simulation Modelling and Analysis ... l2 Project Management Software ...... 32 2.6.1 The basic phases of construction process simulation ...... 12 5 Management Issues 2.6.2 Building a CYCLONE model...... 12 2.6.3 Experimenting, analyzing, and 5.1 Introduction ...... 35 simulating ...... 14 5.2 Quality of Supervision ...... 35 2.6.4 Simulation and productivity ...... 15 5.3 Material Management...... 35 5.3.1 Material management steps ...... 36 3 Human Factors and 5.3.2 Responsibilities ...... 36 5.3.3 Interfaces and their implications Productivity Improve· for productivity ...... 37 ment 5.3.4 Preplanning ...... 38 3.1 Introduction ...... 17 5.3.5 Material control ...... 38 3.2 Motivation ...... 17 5.3.6 Procurement...... 39 3.2.1 Motivation and the construction 5.3.7 Material handling ...... 39 industry ...... 17 5.4 Constructability ...... 40 3.2.2 Factors affecting motivation ...... l8 5.4.1 A traditional problem ...... 41 3.2.3 Motivators ...... 19 5.4.2 Constructability concepts ...... 41 3.2.4 Demotivators ...... 19 5.5 Change Management ...... 4 J 3.2.5 Absenteeism and turnover ...... 21 3.3 Human Factors Related to 6 Conclusion Productivity ...... 21 6.1 Macro- Versus Micro-Productivity . .43 3.3.1 The individual as a factor 6.2 Miscellaneous Ideas for Improving affecting productivity ...... 21 Productivity in Construction ...... 43 3.3.2 Physical limitations ...... 22 3.3.3 The learning curve ...... 22

PRODUCTIVITY IN CONSTRUCTION

1 Introduction

1.1 Productivity: More struction site. A Construction Industry Develop­ Achievement per ment Council task force developed a question­ naire of factors impairing construction Resource productivity (CIDC, 1984). It lists seven cate­ Economists have been saying it, so gories and 95 factors. Table 1.1 lists the most se­ have constructors, organized labour- every­ rious factors within each of the seven categories. body: to remain competitive, we have to Research findings by social scientists produce more for each dollar spent on con­ and construction researchers can be con­ struction. And "we" is everybody - every tentious, due to the difficulty in accounting for worker at a job site can contribute to improved the many interdependencies. The impact of productivity. such factors as morale and satisfaction may be Productivity issues can be divided debatable, but that should not keep us from into macro- and micro-level. At the macro­ thinking seriously about improvements in pro­ level, one deals with contracting methods, ductivity. (The idea of improved productivity labour legislation, and labour organization; at is too important to be allowed to stumble over the micro-level, with the management and op­ academic arguments.) Although we may not eration of a project, mainly at the job site. know the precise effects of many of these fac­ tors, we can observe the effects of combina­ To improve productivity, we must be tions of them. able to measure it. And we must be able to measure the effect of changes adopted on 1.2 What is Productivity? methods, effort, and systems. The measured values of productivity can then be compared Many terms are used to describe pro­ either to those used to compile the estimate or ductivity in the construction industry: perfor­ to some production standards. Although no mance factor, production rate, unit person-hour formal industry standards exist in North Amer­ (p-h) rate and others. Traditionally, productivi­ ica, many sources of published productivity ty has been defined as the ratio of input/output, data, as well as the databases of various com­ i.e., the ratio of the input of an associated re­ panies, can serve as production standards. source (usually, but not necessarily, expressed in p-hs) to real output (in creating economic A number of complex and interdepen­ value). To restate this definition for use in the dent factors can influence productivity on a con- construction industry: labour productivity is the physical progress achieved per p-h, e.g., p-hs Table 1.1 factors seriously impairing construction per linear metre of conduit laid or p-hs per productivity cubic metre of concrete poured. Category Factors The two most important measures of labour productivity are: Project Conditions Weather variability • the effectiveness with which labour is used Market Conditions Material shortages in the construction process; Lack of experienced design and project • the relative efficiency of labour doing what management personnel it is required to do at a given time and place. Design and Procurement Large number of changes Examples of the first measure are the Construction Management Ineffective communications labour dollars required to produce a square Inadequate planning and scheduling metre or square foot of living area, or the labour cost of providing one bed in a hospital. Lack of sufficient supervisory training Another example is the labour content required, Labour Restrictive union rules per barrel of output, to build an oil refinery. In Government Policy Slow approvals and issue of these cases, technological innovations or design permits improvements have the most significant impact Education and Training Lack of management training for super­ because it is the effectiveness with which vision, project management labour is used in the building process that is being measured.

PRODUCTIVITY IN CONSTRUCTION Contractors and organized labour are, of and the causes behind this decline, the con­ however, more interested in the second mea­ struction industry is making strides toward im­ sure, the relative efficiency of labour. Exam­ proving productivity. ples include the number of square metres of formwork or linear metres of conduit that can 1.3 Framework for Produc­ be installed per p-h at a given time and place. tivity Improvement in Labour efficiency is the basis of most Construction tender estimates, as well as the yardstick by which performance is measured and monitored. Productivity improvement in construc­ tion is best understood when the construction For example, it was reported that process is visualized as a complete system as 837.4 p-hs were required to construct a house shown in Figure 1.1. The system is made up of in 1930. By 1965, the requirement was re­ the construction project to which material, per­ duced to 283 .2 person-hours. The reduction in sonnel, equipment, management, and money p-hs is equal to an impressive average annual are inputs. They are consumed by the system growth rate of 3.2%. in the process of producing the construction unit. Control of the system is achieved by col­ It is not surprising that some analysts lecting and processing information about the have tried to explain this as the result of steadi­ rates at which production is attained. ly improving labour efficiency. The real im­ provement, however, had little to do with To measure input/output, the parame­ improved efficiency but was due to such tech­ ter defined as productivity, two types of input nological changes as improved construction ex­ to the system are used: the person-hour/unit and cavating equipment and the introduction of the cost/unit. The first focuses only on labour drywall to replace wet plaster. and is used for labour-intensive operations. The second, cost/unit, combines all effects. When relative growth in labour pro­ The productivity of an operation is measured ductivity was equated with real improvement in and compared to the values in the estimate or labour efficiency, the construction industry was budget. led to believe that no problem in declining pro­ ductivity existed. Apparently lack of motiva­ If the actual productivity does not tion is not seen as a problem, and the ever compare favourably with the estimated values, frequent financial losses were blamed either on the input categories affecting productivity in poor estimating or on the impact of accelerated the system- namely material timeliness, labour schedule performance. Construction supervi­ effectiveness, and management practices­ sors eventually had to face up to reality and need to be examined. admit that labour efficiency has been steadily declining for some time. By accepting the real­ To improve labour effectiveness, vari­ ity and trying to understand both the magnitude ous factors can be addressed, including motiva­ tion, job safety, environmental factors, and physical limitations. Management practices in­ Figure 1.1 Framework for productivity improvement clude scheduling, planning, data collection, job analysis, and control. Material timeliness is en­ Control Procurement sured by proper procurement scheduling, site Scheduling Data Collection layout, and other issues. Planning 1.4 Organization of this Management Publication Practices \ System The purpose of this publication is to Construction introduce the subject of productivity in con­ Project struction. Each topic can be expanded and dealt with in more detail at every level of the \ Labour construction process. The reader can pursue the topics further by referring to the books and papers listed at the end of each chapter. Some of these documents are cited in the text. Unit Motivation The subject matter, aimed at such con­ struction practitioners as project engineers, su­ Take Feedback/Control Compare Actual perintendents or foremen, is presented in a Corrective to Estimated format that can be easily read and understood. Action/Improve It gives practitioners the insights needed to gain an appreciation of productivity in con-

PRODUCTIVITY IN CONSTRUCTION struction. The publication, although an Additional Readings overview of the subject, covers each topic ade­ quately and comprehensively. Adrian, J.J. 1987. Construction Productivity Improvement. New York: Elsevier Science Acknowledgment Publishing Company. The authors have drawn freely from Alfeld, L.E. 1988. Construction Productivity, written contributions to a seminar series jointly On-Site Measurement and Management. New sponsored by the Canadian Society for Civil York: McGraw Hill. Engineering and the Construction Technology Centre Atlantic Inc. The seminars were enti­ CIDC, 1984. Canada Construction Industry De­ tled "Productivity on Canadian Construction velopment Council, 235 Queen Street, Ottawa, Sites: An Overview. What Supervisors Should Ontario K1A OH5. Know and Do." Halpin, D.W. 1985. Financial and Cost Con­ The contributors were C. Fear, R. trol Concepts for Construction Management. French, J. Gibson, G. Jergeas, K. Lemon, S. New York: John Wiley and Sons. Perfect, K. Pressnail, S. Revay, A. Russel, L. Oglesby, C.H., H.W. Parker, and G.A. Howel. Tardif, C. Trembley, L. Waugh, and S.P. 1989. Productivity Improvement in Construc­ Dozzi, who also served as the editor. tion. New York: McGraw Hill. Warren, R.H. 1989. Motivation and Productivi­ ty in the Construction Industry. New York: Van Nostrand Reinhold.

PRODUCTIVITY IN CONSTRUCTION

.. 2 Techniques for Measuring and Improving Productivity at Construction Sites

2.1 Introduction 2.2 Measuring and Inter­ preting Work and Crew The management of site-related issues in construction projects is often complex and Effectiveness difficult. The main problem is the quantifica­ 2.2.1 Field rating tion of all factors involved on site. The most accurate measure of productivity in construc­ Field rating can be used to estimate tion is the number of units produced per per­ crudely the level of activity of a construction son-hour (p-h) consumed, or its reciprocal, the operation. The method simply categorizes the number of p-hs consumed per unit produced. observed worker as either "working" or "non­ The productivity of a process can be measured working" and uses the "working" fraction as a indirectly by observing the level of activity of measure of effectiveness. To collect a random its resources. sample, an observer on site observes the work­ ers. Once a sample has been collected, the field Work studies and surveys may have a rating is calculated as total observations in the demotivating effect on the workforce. Special "working" category divided by the total num­ precautions must be taken to avoid the percep­ ber of observations, plus 10% to account for tion that the company is spying on its workers. foreman and supervisory activity as follows: Education and information sessions are recom­ mended to create a team approach to productiv­ Field rating = total observations of working/ ity improvement. At a micro-level, workers are total number of observations+ 10% a valuable source of information concerning their performance or efficiency. Participation The number should be roughly over by the tradesmen 1 or supervisors can be expect­ 60% for a job to be satisfactory. For example, ed only if requested. if a foreman made 100 observations of workers and only 40 were classified as working at the This chapter deals first with proven time, then the field rating would be 50%, i.e., techniques that are more widely used to mea­ 40/100 + l 0. The job would, therefore, be con­ sure the effectiveness (and, indirectly, the pro­ sidered unsatisfactory. The method does not ductivity) of construction workers and crews. tell the analyzer anything about the sources of Also discussed is the use of the data collected problems or inefficiencies. It merely suggests to improve the productivity of a construction that there is something wrong. process. A discussion on a method for measur­ ing both effectiveness and productivity at the 2.2.2 Work sampling same time, the Method Productivity Delay Work sampling is based on statistical Model, (Adrian and Boyer, 1976) follows. The sampling theory and is a slightly more sophisti­ second part of this chapter deals with more ad­ cated method than field rating. The basic ob­ vanced methods to study and improve the pro­ jective is to observe an operation for a limited ductivity of a given construction process. With time and from the observations infer how pro­ the increased use of computers, simulation is ductive the operation is. Statistical sampling one of the more advanced techniques that could theory is applied because the amount of time improve productivity. The idea of using sys­ tems simulation in planning and analyzing con­ spent collecting data has to be limited. In addi­ tion, the number of workers observed is nor­ struction processes is introduced. To make this mally a small sample taken from the entire as practical as possible, the discussion is limit­ population of possible observations (every ed to the CYCLONE methodology (Halpin and glance at the worker is considered an observa­ Riggs, 1992) that has been developed specifi­ tion and therefore, every work sample can re­ cally for construction operations analysis. sult in a multitude of observations). Instead of dealing with the whole population, the proce­ dure is to collect a sample, analyze it, and build a confidence limit around it.

1The words, foreman, craftsman, and tradesman, are meant in a gender-neutral sense; Work sampling estimates the percent­ likewise the word, workmanship. age of time a labourer is productive relative to

PRODUCTIVITY IN CONSTRUCTION the total time the person is involved in the oper­ Research indicates that the productive ation. To accomplish this, the following ap­ work category should normally be over 30%. proach can be adopted: Results of different work sampling studies vary from the low of 9.4% at Isle of Grain to a high 1. Classify the worker's activity as one of three of 64.4% measured by the National Association modes of activity: productive, semi-produc­ of Home Builders Research Foundation in tive (involved in supporting the main activi­ 1973. Other representative samples reported ty), and non-productive. are: • 32% by the Civil Engineering magazine in Note: There are a number of possible varia­ 1977, measured at various nuclear power tions for this classification and readers can develop their own, once familiar with the sites • 34.7% by S.B. Palmater, measured at 13 nu­ concept. Flexibility can be enhanced by ma­ clear power sites nipulating the semi-productive classifica­ • 46.5% measured by the University of Texas tion. One can easily define various modes at random sites. of semi-productive activity, as shown in Table 2.1. For example, support work can Improving the productivity of the be of the form "material handling," "instruc­ process involves identifying the current activity tion and decision making," "equipment rating and the sources of non-productive or maintenance," and others. However, more semi-productive modes. This can be subjec­ than a handful of classifications can make it tively analyzed and depends heavily on the difficult to collect data on site. level of detail of the classification scheme adopted and the project being analyzed. A 2. Develop a data collection form that will fa­ cilitate tallying the observations on site, as sample set of recommendations is given in Table 2.2. shown in Figure 2.1, for example. 3. Take random observations of workers in­ volved in a given operation in the field. The observation should indicate the workers' ac­ Figure 2.1 Sample work sampling tivity mode, i.e., productive, non-productive, data collection form or semi-productive. Random, for all practi­ Work Sampling Sheet cal purposes, means without any bias as to who is being observed and that each worker Project: will have the same chance of being observed Date: Observer: as any other worker. Notes: 4. Record all observations on the form. Enter a checkmark under the appropriate mode of activity observed. Observation Productive Semi-Productive Non-Productive No. (Direct work) (Support work) (Delay) 5. Add up all the checkmarks under each mode and calculate the percentage of activity. In 1 .J the example (Figure 2.1 ), the 'percent pro­ 2 .J ductive' is calculated as 4/9 (= 45%), the 'percent non-productive' is 3/9 (= 33%), and 3 .J the balance of 22%, semi-productive. 4 .J 5 .,[ 6 .J 7 .J Table 2.1 Examples of activity classification 8 .J Classification Productive Semi-Productive Non-Productive 9 .J (equivalent (Direct Work) {Indirect Work) (Delay) Total 4 2 3 classification) (Working) (Support Work) (Non-Working) Percentage 45% 22% 33% Description Using trade tools Supporting the Not contributing to main activity the activity Examples mason laying brick, tradesman getting personal breaks, labourer mixing material, travelling waiting for equip- mortar, electrician to work location, men! to be fixed, pulling wire, welder taking instructions waiting for more welding pipe instructions, late start or early departure

PRODUCTIVITY IN CONSTRUCTION ------1. Identify the members of the crew to be ob­ Table 2.2 Sample causes of delay and recommended served and structure a form similar to that remedial actions shown in Table 2.3, with the crew to be ob­ served noted in the column headings and the Causes of Delay Suggested remedial action time of observation listed in the rows of the (Excessive percentage of (Each of these actions will require a first column. time spent on this factor) detailed analysis) 2. Observe the crews as they are working. For Waiting for instruction Pre-plan and pre-assign work duties. the observation interval (in Table 2.3 the in­ terval equals 5 minutes), determine whether Finding material Improve site layout. the crew member has been active for over Getting material Examine material handling and site half the interval. If so, mark the observation layout. cell with an "x"; if not, leave the cell empty. Personal breaks Examine human resource management 3. Add the "x" observations for the entire table and discipline. and divide the sum by the total number of Waiting for equipment repair Use stand-by equipment where possible, observations. In the example of Table 2.3, pre-plan to reassign crews to other activ­ 22 observations were positive out of a total ities, schedule equipment maintenance of 32; therefore, the effectiveness is 22/32 or to keep it in good working condition. 68%. Waiting or queuing for service Resolve resource allocation problems, possibly by balancing the resources. Table 2.3 Sample five-minute Material handling difficulties Improve site layout and address safety rating data collection form concerns. Time Spreader Screeder Grader Bull-Floater For work sampling to be effective, the 9:50 X X X observer must make a large number of observa­ 9:55 X X X tions, a number that must be determined from 10:00 X statistical sampling theory. The minimum 10:05 X X X X number generally accepted is 3842 observa­ tions. This number is derived from a sampling 10:10 X X error of 5% and a level of confidence of 95%. 10:15 X X X

Tables, nomographs, and computer programs 10:20 X X X X can be used to calculate the required number of 10:25 X X observations for different sets of error limit or confidence levels. Effective observations 6 6 5 5 Work sampling only attempts to indi­ rectly measure productivity. It is difficult to Total observations= 32 Effectiveness = 22/32 determine the productivity of a carpenter, for Observed effective = 22 5-Minute Rating = 68% example, by observing how many hammer blows it takes to drive a nail. 2.3 Field Surveys In decision-making, the results should The work sampling methods covered be viewed with caution and used with discre­ in Section 2.2 measure efficiencies in the site tion. They cannot be used to measure real operation, but do not go far enough in identify­ labour efficiency, yet they are extremely useful ing the leading cause for the inefficiency. For to gain a better insight into motivation, and at example, work sampling might indicate that a the same time help explain the reasons behind craftsman spent 25% of the time being delayed drastic variations in production rates. because the required material was not available. 2.2.3 Five-minute rating The method cannot, however, pinpoint the real cause of the delay or what can be done to re­ The five-minute rating technique, un­ duce it. like work sampling, is not based on statistical sampling theory. The method relies on simply Field surveys and questionnaires are observing an operation for a short time. The organized ways of involving the foreman or observation does not result in a large enough craftsman in the site evaluation and productivi­ sample to support work sampling. The method ty improvement process. Craftsmen are proba­ 2Aithough this seems to be does, however, provide some insight as to the bly the persons most familiar with their work a large number, in actual activity. They can easily identify sources of applications, every glance effectiveness of the crew and can identify areas at the work in progress is where more observation is required. delay and obstacles in their progress. Likewise, an observation. Therefore, a foreman is the person most familiar with the 384 observations is not The following procedure can be used crew and the problems that restrict improve­ excessive. to implement the 5-minute rating technique: ment in their productivity.

PRODUCTIVITY IN CONSTRUCTION ------2.3.1 Foreman delay survey Figure 2.2 Typical FDS form Foreman Delay Survey (FDS) relies Problem Causing Area Person-Hours Lost on a questionnaire which is to be filled out by No. of No. of Total the job foreman at the end of a working day ac­ Hours Lost Workers Person- Hours cording to a particular survey schedule, e.g., one work week in each month. The question­ Redoing work (design error or change) naire is primarily meant to identify the number Redoing work (prefabrication error) of hours of a day lost due to delays. Most FDSs Redoing work (field error or damage) are divided into rework and delay categories. Waiting for materials (warehouse) Once a form has been filled out, the informa­ tion is extracted in the form of percentages and Waiting for materials (vendor furnished) action taken to ensure that sources of delays are Waiting for tools properly dealt with. A typical FDS form is Waiting for construction equipment given in Figure 2.2. Construction equipment breakdown The results of the survey are converted Waiting for information from p-hs into equivalent percentages andre­ Waiting for other crews ported on a form as shown in Table 2.4. The information on the report sheet will identify Waiting for fellow crew members concerns that the foremen have with the opera­ Unexplained or unnecessary move tion. The example in Table 2.4 reveals that too Other: much time is being spent on redoing work due Comments: to design error - 2.3% of the time - and waiting for construction equipment - 1.1% of the time. The FDS is a relatively low-cost method for analyzing the sources of delay dur­ ing construction. It can be easily stylized and Made by: implemented. For further details regarding im­ Date: plementation of FDS in construction, refer to Tucker et al., 1982. 2.3.2 Craftsman questionnaire Craftsman questionnaire (CQ) is a questionnaire-oriented technique attempting to address issues and concerns that relate to a craftsman's productivity and motivation. The basic idea is to distribute a simple question­ naire, similar to the one shown in Figure 2.3, to Table 2.4 Sample FDS results craftsmen on a job site to complete. The aim is to identify major factors that inhibit the produc­ Problem-Causing Area P-hs Lost Percentage tivity of craftsmen and estimate the p-hs lost per craftsman per week due to specific causes. Redoing work (design error or change) 122 2.3 The questionnaire can comprise 50 Redoing work (prefabrication error) 24 0.5 short questions addressing such areas of con­ Redoing work (field error or damage) 52 1.0 cern as material availability and site layout, Waiting for materials (warehouse) 33 0.6 equipment and tool availability, rework items and causes of rework, management interference Waiting for materials (vendor furnished) 22 0.4 and inspection, and suggestions for improving Waiting for tools 12 0.2 the process. In addition, the questionnaire asks Waiting for construction equipment 56 1.1 for the hours lost per week per craftsman on Construction equipment breakdown 15 0.3 each area of concern listed. This is often sup­ plemented with personal interviews with some Waiting for information 12 0.2 of the craftsmen to validate the responses and Waiting for other crews 14 0.3 test the level of seriousness. Waiting for fellow crew members 10 0.2 Once the questionnaires have been Unexplained or unnecessary move 20 0.4 collected, results are compiled and statistics re­ Other 70 1.3 ported to all concerned in a form similar to Total 462 8.9 Table 2.5. Total work in person-hours 5 210

PRODUCTIVITY IN CONSTRUCTION ------The report in Table 2.5 implies that Figure 2.3 Sample craftsman questionnaire material availability (13%) and redoing work (12%) are areas of major concern: they con­ Personal data Check ..[the appropriate box for YES or NO, tribute 25 % of the time lost by a craftsman in a or fill the box with the required information. week. From the answers in the questionnaire Craft itself, the reasons for the time loss could proba­ bly be identified. Location Type of work The ability to improve the productivity of an operation from the conclusions drawn Other from the CQ greatly depends on how well the YES NO questionnaire is structured, detailed, and styl­ Material ized, and on how serious the craftsmen's partic­ Is material always available when you need it? ipation is. How many hours do you estimate are lost per week due to 2.4 The Method Productivity material not being available? h Delay Model Tools Are tools always available when needed? The method productivity delay model Are tools in acceptable shape? (MPDM) was proposed as a way to combine both time study and productivity measurement Are tools supplied always the right ones for the job? (Adrian and Boyer, 1976). MPDM relies on Are there any specific tools in short supply (please name) having an observer collect data, on a special How many hours do you estimate are lost per week due to form, pertaining to the cycle time of a leading tools not being available or acceptable for the job? h resource on the operation. The observer also Equipment notes the nature of the delays during the period of observation. Once the data collection is Question 1 (Add more questions as under material and tools) complete, a set of computations is carried out Question 2 (Add more questions as under material and tools) that measures the productivity of the operation, How many hours do you estimate are lost per week due to . . . h indicates the major sources of delay, and gives other useful statistics. Rework Question 1 (Add more questions as under material and tools) MPDM can be an effective way of Question 2 (Add more questions as under material and tools) measuring productivity on site and the delays that undermine it. Experience with the tech­ How many hours do you estimate are lost per week due to ... h nique has shown that it can be less confusing Safety Concerns when implemented on an electronic spread­ Question 1 (Add more questions as under material and tools) sheet. For the example presented in this sec­ tion, Microsoft Excel was used. Any Question 2 (Add more questions as under material and tools) spreadsheet can be easily automated and gener­ How many hours do you estimate are lost per week due to .. . h alized with macros, so that the computations Others are automatic, once the observations have been Question 1 (Add more questions as under material and tools) entered. Question 2 (Add more questions as under material and tools) MPDM provides more information than other work sampling techniques. In addi­ tion to providing the user with a measure of productivity, it can also identify sources of Table 2.5 Results from a CQ delay and their relative contribution to the lack of productivity. Problem/Cause P-hs lost Percentage per week per week MPDM consists of the following Material not available or phases: poorly located 5.2 13.0 1 Identification of the production unit, and the Tools not available or production cycle suitable 3.2 8.0 Equipment not available The production unit is defined as a measurable or down for repair 2.0 5.0 amount of work that can be visually identified Work redone 4.8 12.0 by the observer without much effort. Examples of this would be a bucket of concrete, a truck­ Management interference 2.1 5.3 load of dirt, or a row of bricks. The production Other 2.5 6.3 cycle is the total time that it takes the crew to Total 19.8 49.5 place one production unit.

PRODUCTIVITY IN CONSTRUCTION ------~ 2 Identification of the leading resource Table 2.6 Sample type of delays identified during The leading resource is that resource involved MPDM data collection in the operation with the most impact on the productivity. In other words, the operation will Environmental Equipment Labour Material Management come to a halt if the resource stops producing. Change in soil Equipment Personal break Not available Poor planning An example would be a crane in a concrete conditions being positioned when needed placement operation, the mason in brick-laying, Change in Temporary Finding Defective and Undecided as or a dozer in an earth moving operation. This wall section breakdown materials has to be to what should resource will be the centre of observation and or tools replaced be done cycle time determination. Unscheduled Getting Improperly Unavailable for maintenance instructions located on site instructions 3 Identification of the types of delay that can be encountered in the process Late arrival, Interfering with early departure other operations Five possible types of delay include those caused by environment, equipment, labour, material and management. Experience shows that users should define their own types of delay. Figure 2.4 MPDM data collection sheet 4 Data collection MPDM Data Collection Sheet MPDM requires that the observer time the pro­ Date: June 6, 1992 duction cycle for each production unit placed. Operation: Roof truss installation Observer: SMA The observer must also determine whether a Production unit: One truss Unit of time: Second delay took place during a given cycle. If a delay occurs, the observer must indicate its na­ Prod. Cycle Enviro. Equip. Labour Mat. Mngt. Processing Cycle Time Delay Delay Delay Delay Delay column* ture based on the categories of delays given in Phase 3. Examples of the types of delay under (1) (2) (3) (4) (5) (6) (7) each category are given in Table 2.6. 1 354 12.83 5 Data processing, model analysis and recom­ X 98.17 2 465 mendations 3 343 23.83 The processing of the MPDM data is carried 78.17 4 445 X out by filling out the MPDM data collection 5 504 X 137.17 sheet (Figure 2.4), and Tables 2.7 and 2.8, 6 470 X 103.17 which are meant to be self-explanatory. First, 7 395 28.17 column (7) in Figure 2.4 has to be completed. This is simply column (1) minus the average 8 345 21.83 cycle time of cycles where no delay has oc­ 9 360 6.83 curred. This is also given in Table 2.7, column 10 400 33.17 (3) for a non-delayed production cycle. 11 460 X 93.17 The observer can use the form given 12 385 18.17 in Figure 2.4 to facilitate the data collection. Sometimes MPDM fails to work because the 13 360 6.83 cycle time is too short to observe, or too long to 14 353 13.83 keep track of. In such cases, the method is not 15 372 5.17 recommended. If time-lapse film is available, 16 505 50%** 50% 138.17 short processing cycles can be captured. 17 465 X 98.17 To illustrate the data collection proce­ 18 440 X 73.17 dure, consider a simple process involving the installation of a roof truss. The production unit 19 430 X 63.17 identified for the roof truss process (Figure 2.4) 20 360 6.83 was the actual placement of a wooden roof 21 375 8.17 truss. The production cycle began with the lift­ ing of one truss and concluded when that same 22 405 X 38.17 truss was permanently braced. The leading re­ 23 475 X 108.17 source was the mobile crane used to place the * This column is not part of the data collection. It is inserted in the Table to facili­ truss members. The cycle times were timed by tate processing. reviewing the time-lapse film of the operation. To fill out this column take column (1) minus the average of the cycle times Potential sources of delay were also where no delay occurred. recorded whenever noticed. If more than one ** To attribute delay to more than one source, use percentages. delay is observed during the same cycle, then the share attributed to the delay should be noted

PRODUCTIVITY IN CONSTRUCTION ______..

as a percentage (Row 16 in Figure 2.4, for ex­ Table 2. 7 Summary of MPDM computation ample). On the first observation of the crane's cycle time, it took 354 seconds to place one Production Number Mean Cycle I[ICycle truss; no delays were observed. The entry in Total Time of Cycles Time time-Non-delay Figure 2.4, Row 1 is recorded under Col. 1. cycle timel]/n On the second cycle, which took 465 seconds, (1) (2) (3) (4) an equipment delay was noticed. The entry in Non-delayed Sum of all cycles No. of cycles Col. 1 + Col. 2 Col. 7 from Row 2 Col. 1 records the cycle time, and an "x" production (in Col. 1 of where no delay Fig. 2.4 for was entered under Col. 3 to indicate an equip­ cycles Fig. 2.4) where occurred · non-delayed ment delay. The remainder of the data collec­ no delay was cycles .;. Col.2 tion form was filled in similarly. observed Upon completion of the calculations Overall Sum of all cycles Total number Col. 1 .;. Col. 2 Col. 8 from performed according to Table 2.8, the MPDM production (sum of Col. 1 of of cycles Fig . 2.4 for equations can be used to compute the produc­ cycles Fig . 2.4) non-delayed tivity of the operation. cycles + Col.2 Overall Method Productivity =(Ideal Productivity)(l - 2: expected % of delay time) Table 2.8 MPDM delay information Note: 2: expected % of delay time is the sum of Row e in Table 2.8. Time Variance Environment Equipment Labour Material Management Ideal Productivity (1) (2) (3) (4) (5) = 1/Mean cycle time for non-delayed cycles No. of occurrences Total x's in Col. 2, Figure 2.4 for this type of delay Note: Mean cycle time for non-delayed cycles Total added time Sum of this type of delay from Col. 7, Fig. 2.4 is obtained from Row a, Col. 3 in Table 2.7. Note: When percentages are used in Fig. 2.4, then the sum should be prorated to the source of delay. The last step in the MPDM computa­ tions requires the development of the variabili­ Probability of occurrence Row a +total number of cycles ty of the ideal and overall production rates. Relative severity Row b + Row ax mean cycle time for overall cycles, i.e., These rates must first be analyzed to assess the Row b, Col. 3 of Table 2.7 variability of method productivity. Adrian and Expected percentage of Row c x Row d x 100 Boyer (1976) state that the higher the overall delay time cycle variability and the ideal cycle variability, the less dependable the productivity prediction. Ideally, these ratios should be small. The vari­ ability of the productivity indicators are calcu­ Table 2.9 MPDM processing for sample applications lated from Table 2.7 as follows: Ideal cycle variability = Value of Row a, Col. 4 Prodution Number of Mean Cycle I.[ICycle Time-Non- -;- value from Row a, Col. 3 Total Time Cycles Time Delay Cycle Timel]/n (1) (2) (3) (4) Overall cycle variability= Value of Row b, Col. 4 -;- value from Row b, Col. 3 Non-delayed 4 402 12 366.83 15.47 production cycles To illustrate the computations for Overall 9 466 23 411.57 52.81 MPDM, an example is given in Figure 2.4. production cycles The cycles that were identified as 'non-de­ layed' totalled 12 cycles with an accumulated cycle time of 4 402 seconds. The result is a mean cycle time of 366.83 seconds (6.1 min­ utes). The processing is then performed and Table 2.10 Delay information for sample the results are shown in Tables 2.9 and 2.10. Time Variance Environment Equipment Labour Material Management The ideal productivity would then be (1) (2) (3) (4) (5) (60 hours/min)/(6.1 minutes/cycle)= 9.81 trusses/hour. The real productivity, however, is No. of occurrences 2 5 3 the ideal productivity adjusted for the expected Total added time 141 .3 440.8 69.1 137.2 240.4 percentage of delay time per production cycle. Probability of The real productivity is calculated to be 8.75 occurrence 0.09 0.22 0.04 0.04 0.13 trusses per hour. Relative severity 0.17 0.21 0.17 0.33 0.19 The variability rates would be: Expected percentage of delay 0.01 0.05 0.01 0.01 0.03 Ideal cycle variability= 15.47/366.83 = 0.04 (or 4%)

PRODUCTIVITY IN CONSTRUCTION --Overall----- cycle variability= 65.281/411.57-- = --tivity times for each element may be deter­ 0.13 (or 13%) mined during one cycle. Conversely, the use of a stopwatch requires sampling from many Such variability is considered to be cycles in order to record the activity times of relatively minor and would indicate that the each crew member. It is best to show only productivity rates obtained are realistic values. those elements that are pertinent to the problem The results of the analysis can be used at hand, because a crew balance chart may be­ to determine what the productivity rate is and come cluttered with useless information, which what can be done to improve the productivity. reduces its effectiveness. Table 2.10 indicates that the most expected From a crew balance chart, the user delay can occur from equipment (5%), but the may detetmine interrelationships by comparing most severe (lengthiest) delay has occurred activities along a horizontal line since the time from material. The 22% probability of an scale is the same for each crew member. In this equipment delay indicates that management manner, inefficient crew size or organization is should concentrate on solving the problems as­ identified and remedial action can be taken. By sociated with equipment. analyzing a crew balance chart, the user is stim­ ulated to devise more efficient methods of per­ 2.5 Charting Techniques: forming the task. Reorganization of the crew Crew-Balance Charts may be all that is required or a different method may be in order. Crew balance charts are a method of comparing interrelationships among various 2.6 Simulation Modelling crew members and equipment required to can)' and Analysis out a task. This method is applicable to such cyclical tasks as placing concrete. Simulation in the context of this dis­ cussion is defined as "building a mathematical/ Vertical bars, as shown in Figure 2.5, logical model of a system and experimenting represent each person or machine element in­ with it on a computer" (Pritsker 1986). This volved in the task at hand. The ordinate of the publication addresses simulation only with re­ chart expresses time either as a percentage of gard to the CYCLONE methodology. Al­ the total cycle time or the actual time of day. though other techniques exist, none has shown Each bar is subdivided vertically to show the so much promise in construction as the CY­ time required for each activity involved in the CLONE methodology. task cycle, including idle, non-productive, and ineffective time. 2.6.1 The basic phases of con­ struction process simulation To construct a crew balance chart, the time for each activity in the cycle is recorded The simulation process consists of two for every worker or machine involved in the basic phases: modelling and experimentation. task. This may be done using a stopwatch or CYCLONE provides the modelling elements time-lapse film. The use of time-lapse film has and methods that a modeller can use to repre­ many advantages over a stopwatch in that ac- sent a construction operation in much the same way as a scheduler would build a Critical Path Figure 2.5 Sample crew balance chart for a concret· Method (CPM) network for a construction pro­ ing operation (Cycle time = 4 minutes} ject, i.e., by specifying activities, and their logi­ cal relationships, durations and resource Crane Bull Float Vibrator 1 Worker Spreader 1 requirements. To model an operation using 3.5 Bucket ready 9 Idle CYCLONE, the modeller focusses on there­ 16 Fill bucket 20 Work sources involved and their interactions. Are­ Work 34 Wait 2.5 Bucket idle 26 Work source can be in one of two states: active or Move bucket 8 15 Idle idle. An active state of a resource is represent­ 10 Pour concrete ed by a square; the idle state, by a circle. In the 'if< 16 Work a) 10 Move bucket 25 Idle 20 Work model, the resource can move between the two u Idle >- 3.5 states and thus from one activity to the other. u Fill bucket The whole idea of simulation revolves around 16 with concrete 15 Idle 33 Wail the dynamic movement of resources. It is es­ 2.5 Bucket on lloor Work Move bucket 25 8 back to floor 12 Work sential to distinguish between this method and a

10 Pourcoocrete 28 Work static system like CPM. Idle Move bucket to 15 Idle 18 17 Work 10 pour concrete 2.6.2 Building a CYCLONE model 88 72 51 52 33 A CYCLONE model is constructed by Time working (%) using the CYCLONE elements shown in Table Effectiveness, Total cycle time 2.11.

PRODUCTIVITY IN CONSTRUCTION ------The rules for structuring CYCLONE 4. Establish the logical relationships between network models using these elements are sum­ these tasks (i.e., precedence and sequencing marized in Table 2.11. of the tasks) by connecting the COMB I, NORMAL, and QUEue nodes with direc­ The CYCLONE modelling procedure tional flow arrows indicating where the re­ uses the following steps: source would be moving from and to upon 1. Identify all resources involved in the opera­ completion of a task. This makes up the tion to be modelled. CYCLONE network. 2. Define the tasks (active states of a resource) A simple example of a CYCLONE composing the process to be modelled. Rep­ network model of an earth-moving operation is resent them with CYCLONE square elements given in Figure 2.6. A stockpile of dirt has to (a task that is constrained by the availability be moved from one location to another. The of more than one resource is represented by a dirt would have to be loaded into hauling units COMBination element and a non-constrained first, the hauling units would transport the dirt task by a NORMAL element). to the required location where it will be dumped. After dumping its load, the hauling 3. Define the resource requirement in the tasks unit returns for another load. and decide where they should wait when a constrained task is not available for service, i.e., it is waiting for other resources before it Figure 2.6 CYCLONE model of an can proceed. This defines circle elements earth-moving operation known as QUEue nodes in CYCLONE ter­ minology.

Table 2.11 Rules for structuring CYCLONE models

CYCLONE Description and Rules for Model Building Element NORMAL The NORMAL is not a restrained task. Any resource that arrives at a NORMAL is given access and is immediately processed. It is like a serving station with an infinite number of servers. Can be preceded by all other CYCLONE elements except for a D QUEue node. Can be followed by all other elements except for a COMBI. COMB I A task that is restrained by the availability of more than one type of resource. A resource arriving at a COMBI will have to wait until all other required resources are available before it is given access to the task. D Can be preceded by QUEue nodes only. Can be followed by all other elements except COMB Is. QUEue A QUEue node is a waiting area for a resource. Therefore it is used only when a task is restrained. A resource arriving at a QUEue node will stay in the node until a COMB I is ready to process it. A QUEue node has one other function in the MicroCYCLONE imple­ 0 mentation, namely to multiply resources when specified. In other Assume that this simple operation will words, a modellor can specify that once a resource enters a speci­ be accomplished by Of!e front-end loader (FEL), fied QUEue node, it will multiply into a finite number of duplicate re­ three trucks, and one labourer to spot the dump sources. location. This completes Step 1. In the previ­ Can be preceded by any element except a QUEue node. ous paragraph, the words that describe the tasks Can be followed by COMBis only. required to complete the operation have been FUNCTION The FUNCTION element was devised to provide some flexibility. Dif­ italicized. This completes Step 2. The equiva­ ferent computer implementations of CYCLONE have somewhat dif­ lent CYCLONE elements are then matched with ferent functions. In MicroCYCLONE, one type of function is allowed, the proper task and arranged as shown in Figure namely the consolidate function. Its job is to take units and consoli­ 2.6. The loading task was restrained by the 0 date them into a specified number. Any unit arriving at this function will accumulate until a threshold value is reached, at which point only availability of both the truck and the FEL and one unit is released from the function (all others are destroyed). therefore it is modelled by a COMBI node (sim­ Can be preceded by all elements except QUEue nodes. ilarly the dumping task, which requires the truck Can be followed by all elements except COMBis. and the spotter, was modelled with the COMBI COUNTER The counter keeps track of the number of times units pass it. It does node). The travel to dump location requires the not alter any of the resources or their properties. It just adds incre­ truck only and therefore was modelled by a ments and keeps track of cycles and a few other statistics. NORMAL node. The truck-retuming task was Can be preceded by all elements except QUEue nodes. a NORMAL node for the same reason. When Can be followed by all elements except COMB Is. the task was restrained, it is preceded by two QUEue nodes where the respective resources

PRODUCTIVITY IN CONSTRUCTION wait. Loading was preceded by the FEL productivity or achieve minimum unit cost, as QUEue where the PEL waits until a truck is well as to deal with uncertainty and risk. available. The dirt was modelled with a QUEue The operation considered in this model node and the truck waiting for loading also with involves earth moving for a sports training facili­ a QUEue node. The truck waiting for the ty in a university area. The earth is to be moved dumping was modelled by another QUEue from the location of the training facility to the node. This emphasizes that the state of the re­ dump location about 3 km away. Two excava­ source is modelled in CYCLONE methodology, rather than the resource itself. tors remove earth at the job site at two different locations. A number of trucks carry the dirt, 2.6.3 Experimenting, analyzing, dump it, and return for another cycle. The truck and simulating would normally wait until one of the excavators is freed up before proceeding for loading. The Once a model has been built, it can be operation was observed and data collected on the entered into a computer program such as Mi­ cycle times of the various equipment using a croCYCLONE for processing and performing stopwatch. The observer also noted that the the simulation study. The results of the simula­ trucks break down on almost 5% of the cycles tion study are: due to overloading (e.g., from flat tires). o an estimate for completing the operation o the hourly production rate The CYCLONE model of the opera­ o other measures of equipment utilization. tion was prepared as discussed previously. One main difference is the modelling of truck break­ Figure 2.7 presents another model of down. The branch corning out of dummy an earth-moving operation. It is somewhat dif­ NORMAL node (No. 14) indicates that every ferent from the one in Figure 2.6. The model time a truck passes this task, it has 5 chances in can be used to balance resources, maximize 100 of ending up being tied in the repair task (No. 8); 95 times in 100 it continues in its Figure 2. 7 Another CYCLONE model of an earth· cycle. Upon repair, it is released back to its moving operation original cycle. This is how equipment break­ downs are modelled in CYCLONE to simulate Excavator idle actual breakdown in the operation. at position 1 Now the model can be entered into Mi­ croCYCLONE. The MicroCYCLONE user manual (Halpin, 1990) gives the reader all the required information. The first step is to transfer the graphical model into a written text file in the MicroCYCLONE syntax. The model in Figure 2.7 translates into the file shown in Figure 2.8.

Figure 2.8. Simulation input file NAME 'Earth-Moving' LENGTH 5000 CYCLE 100 NETWORK INPUT 1 COMB I SET 'LOAD @ 1'FOL 2 3 PRE 2 10 2 QUE 'EXCAVATOR1 IDLE' 3 NOR SET 2 'TRK BACK CYC' FOL 6 4 COMB I SET 3 'LOAD @ 2'FOL 3 5 PRE 5 10 5 QUE 'EXACVATOR2 IBLE' Excavator idle at position 2 6 FUN COU FOL 14 QUA 1 7 QUE 'TRK QUEUE' 8 COMBI SET 4 'TRUCK REP' FOL 910 PRE 7 9 9 QUE 'REPAIR CREW' 10 QUE SEL LOAD POSITION' 14 NOR SET 5 'TRK BREAKDOWN' FOL 7 10 PRO 0.05 0.95 SEED 101 RESOURCE INPUT 1 'EXCAVATOR' AT 2 FIX 129.38 4 'TRUCKS' AT 10 FIX 50.86 1 'EXCAVATOR' AT 5 FIX 129.38 1 'REPAIR CREW' AT 9 FIX 28 DURATION INPUT Truck queue SET 1 5 SET2 35 SET3 9 SET 4 60 Repair crew idle SET 50 ENDDATA

PRODUCTIVITY IN CONSTRUCTION ------C»

This is entered into MicroCYCLONE highest productivity in truck-loads/hour is and the simulation is started from the program. about 16 trucks. The best cost/unit value is for We started with four trucks working with the 12 to 16 trucks and jumps slightly with 10 and two dozers. The program outputs a productivi­ 18 trucks. ty curve as shown in Figure 2.9. The possibilities of analysis are almost Now a multiple simulation scenario is endless, once the simulation model has been run to find the best combination of trucks and constructed. The modeller can, for example, dozers. The assumption is that two bulldozers try different combinations of equipment, check and 25 trucks are available. The simulation is what happens to the system if the dump loca­ performed and the results of production per tion is changed, estimate the time required to hour versus number of trucks as well as the move a specific quantity of dirt, and so on. cost/unit versus the number of trucks are given in Figure 2.10. The combination yielding the 2.6.4 Simulation and Productivity Simulation can be a very effective tool Figure 2.9 Productivity curve for the simulated to plan for productivity. Moreover, simulation process studies have been conducted to understand bet­ ter the effect of various factors on productivity. 8 Simulation can also be used to support claims Productivity (trucks/hour) vs cycle number due to loss of productivity from bad weather, 6 unexpected delays, changed conditions, changes in the contract, and other factors. Sim­ ilar studies can be conducted to analyze the ef­ fect of particular human factors on productivity. The construction industry is far more 0 complex than the service and industrial sector. 11 21 31 41 51 61 71 81 91 Construction projects often take place in an Cycle Number open environment that changes with every pro­ ject. Repetition is not obvious and the work­ force is diversified. Construction is a unique industry and we should view it as such when Figure 2.10 Summary of the simulation results we examine a technique used for managing it. The methods described here have been tried on numerous projects. Unfortunately, the construction industry, which is traditionally Unit cost vs number of trucks 2 craft-oriented, has not taken the steps required () 150 2 to use more advanced tools in its attempts to ~ improve productivity. Initiatives to use new Cii 100 methods and techniques for measuring and im­ 0 () proving productivity should be taken at the in­ 50 dividual level.

4 6 8 10 12 14 16 18 20 22 24 Number of trucks

25

2 Production vs number of trucks en 20 --~ () ~ 15 c: 0 10 n:::J '0e c.. 5

0 2 4 6 8 10 12 14 16 18 20 22 24 Number of trucks

PRODUCTIVITY IN CONSTRUCTION ------~ ------Additional Readings Adrian, J. and L.T. Boyer. 1976. "Modeling MeU10d- Productivity." ASCE J. Const. Div. 103 (3):154-168. Halpin, D.W. 1977. "CYCLONE: Method for Modeling of Job Site Processes." ASCE J. Const. Div. 103 (3):489-499. Halpin, D.W. and R.W. Woodhead. 1976. De­ sign of Construction and Process Operations. New York: John Wiley and Sons. Halpin, D.W. 1990. MicroCYCLONE User's Manual. Division of Construction Engineering and Management, Purdue Uni versity, West Lafayette, Indiana. Halpin, D.W. and L.S. Riggs. 1992. Planning anti Analysis of Construction Operations. New York: John Wiley and Sons. Pritsker, A. J 986. Introduction to Simulation and SLAM II. New York: John Wiley and Sons. Tucker, R.L., D.F. Rogge, W.R. Hayes, and F.P. Hendrickson. 1982. "Implementing Fore­ man Delay Surveys." ASCE J. Const: Div. 108 (4):577-591.

PRODUCTIVITY IN CONSTRUCTION 3 Hu111an Factors and Productivity l111prove111ent

3.1 Introduction 3.2 Motivation The motivation of workers can be en­ When applying motivational theories hanced through job enrichment (increasing the to everyday problems, three questions arise: things that satisfy workers about a job) and by • What energizes human behaviour? lessening the demotivators (the things that • What directs such behaviour? workers dislike). Reducing demotivators only, • How is this behaviour sustained? the predominant practice of North-American The answers given by various social management, is not enough; it should be sup­ scientists have been expressed in different plemented with job enrichment. words; nevertheless, they all seem to agree that Workers are motivated by completing human beings are energized by their physiolog­ productive quality work, creating or building ical needs, and that their behaviour is directed something, and social relations at work. Pro­ by their expectations and sustained by obtain­ ductive work can be produced by good plan­ ing just reward. ning and communications. Satisfactory social 3.2.1 Motivation and the construc· relations are simply working with other work­ tion industry ers who are friendly and respectful. Individuals and organizations need goals to try to meet or The July 10, 1980 issue of The exceed. Workers can often be motivated Listener, a magazine published by the British through goal-setting. Goals must be clearly es­ Broadcasting Corporation, contained an article tablished to elicit maximum performance and describing the British experience in construct­ provide a feeling of maximum individual ing nuclear power plants. The following is an achievement. Individuals need a system or excerpt from that article: method by which to measure their achieve­ ments and compare their standings against a "In recent years, no big plant has been put up given target. on time. No big plant now being built is on schedule. The delay ranges from two to two Construction work is varied, which and a half years for a chemical plant to four can be satisfying. Workers are often motivated years for something as big as Grain (Nuclear because they see the progress and results of Power Plant, Isle of Grain, G.B.). The rot is their work. There are also many demotivators. not solely or even mainly due to strikes; it is The most common include: the result of almost unbelievably low levels • non-availability of the right material, tools, of productivity. Do you believe that a man or equipment can spend eight hours on a site but do only • poor relations between workers and man- 45 minutes' work in the whole day? agement No? But he can, and this is how they do it. • poorly organized projects In a standard eight-hour day, clocking on, • breakdown in communication walking to and from the job, tea-breaks, bad • lack of recognition of outstanding efforts weather, union business, leave less than four • disrespectful treatment hours available for actual work. Inefficiency, • unfair work assignments overmanning, and other bad habits will eat • incomplete engineering/design work into another two or more hours, and you are • lack of cooperation between different crafts left on a good British site with, at best, one • poor supervision hour and 40 minutes of actual working-time. • rework On a bad site, where ten-minute tea-breaks • no participation in the decision-making have been known to stretch to an hour, the process figure comes down to 40 minutes. Shop • restrictive or burdensome procedures. stewards can tell tales of awkward jobs that Worker satisfaction and motivation take hours to set up or high chimneys that can be increased by removing or reducing these take half an hour to climb up and thus ex­ problems. Questionnaires of the type described plain away the little time a man spends with in Chapter 2 and suggestion boxes can be use­ tools actually in his hands. There is, how­ ful in bringing these problems to the surface. ever, only one such chimney at the Grain.

PRODUCTIVITY IN CONSTRUCTION ______.,

These are all actual, audited figures and they 3.2.2.1 Planning reflect deterioration; productivity at the Grain is about 30 percent worse than it was at two Planning includes both overall job or­ earlier and comparable power stations ganization and work distribution at site level. built...." Higher-level planning must provide for effi­ cient sequencing of the various phases, e.g., The technique used to measure 'pro­ design must precede the preparation of con­ ductivity' at the Isle of Grain site was 'work struction drawings and on-site construction sampling' (see Chapter 2). One measures should not start until adequate drawings are through random observations the ratio of pro­ available. Similarly, subsequent trades should ductive time to total available time. For the not be called to the site until the preceding purposes of such studies, productive time is de­ trade has made enough progress to allow an un­ fined as the time spent on cutting material, interrupted work flow. Site management must hoisting equipment, installing components, or ensure that required material is available in suf­ erecting formwork, in general, working with ficient quantity for continuous progress. Good tools in hand. planning motivates workers because they can build up and maintain momentum toward com­ To understand the influence of moti­ pleting their assigned task without interruption. vation one must analyze all factors that affect efficiency. Labour efficiency is the rate at Good planning practices include proper which workers do what they are required to do use of scheduling techniques, site-layout plan­ at a given time and place. To the extent that ning, procurement scheduling, work assignment the terms 'labour efficiency' and 'labour pro­ and organization, and proper approaches to eli­ ductivity' are used interchangeably, labour pro­ sis management. Good planning also involves ductivity in this context means the rate of feedback and control mechanisms. (For further physical progress of a single task per p-h, information about these planning issues, con­ where the added value has resulted from the sult such project management textbooks as input of human efforts only. In simplistic Ahuja, 1984; Halpin and Woodhead, 1976; and terms, labour efficiency is governed by both Hendrickson and Au, 1989.) workers' attitude toward their assigned task and 3.2.2.2 Communication their ability to perform it. Unfortunately, this definition tends to put the entire responsibility To be able to contribute to the success for efficiency, or the lack of it, onto the shoul­ of a project, a worker must be told exactly what ders of labour, which is obviously wrong, be­ tasks are expected of him. Therefore clear ex­ cause management has as much, if not more, planations of tasks and expectations are required. control over efficiency than labour has. Employees must also know where their instruc­ tions come from, i.e., there must be a visible The factors that control labour effi­ communication chain on the job. Instructions ciency include extraneous constraints, such as from an unknown source will be disregarded. governmental regulations, climatic conditions, Moreover, to be totally successful, the communi­ union rules, skill or inherent attitude of labour, cations should flow both ways. The 'bottoms­ and management practices. up' management system practised in Japan does improve productivity. The system works be­ Motivation is divided into two compo­ cause it nurtures the idea of communicating nents, namely: ideas both upward and downward. • Attitude possessed by the individual when arriving at the site. This attitude may be Instructions and drawings are two the result of the individual's social back­ methods of communication. Each must be com­ ground, family relations, religion, or even plete and timely to allow good planning. Re­ political affiliation. cent developmemts in scheduling and control • Motivation resulting from the various job­ software allow stylizing reports for individuals. related factors controlled by management. In other words, a foreman in charge of form­ work can get a report which only addresses ac­ 3.2.2 Factors affecting motivation tivities of concern to his or her particular line of work and responsibilities. Foremen can thus Based on the experience gained from focus on the required resources, and the start work sampling studies, management practices and progress of each activity for which they are that affect motivation can be good planning, ef­ responsible. This can greatly enhance the in­ ficient communication, and a good work envi­ structions provided to personnel responsible on ronment. Cleanliness, safety, adequate sanitary site and enhance the communication process. facilities, protection from inclement weather, fair but firm discipline, and provisions to ap­ 3.2.2.3 Work environment portion and distribute just rewards are the at­ Creating the proper climate for the tributes of a good work environment. motivation of construction workers depends to a great extent on the attention given to the basic

PRODUCTIVITY IN CONSTRUCTION personal comforts. This issue, when neglected • increase the accountability of individuals by management, can be devastating to the atti­ for their own work tude of labour and becomes a demotivator. • give a person a complete natural unit of work Basic personal comforts on a construc­ • grant additional authority to an employee tion job represent the conveniences which a in his or her activity human being has come to expect in today's en­ • make periodic reports directly available to vironment. They should include drinking the worker water, proper sanitary facilities, site access, • introduce new and more difficult tasks parking and protective gear. • assign individuals specific or specialized 3.2.2.4 Discipline tasks, enabling them to become experts. In addition to remaining alert for un­ The principles can be applied at any even application of the rules, a manager should level on a construction site. be prepared to recognize and praise exemplary 3.2.4 Demotivators performance. Failure by supervisors to enforce discipline or take corrective action can demoti­ Demotivators have only a negative ef­ vate the entire labour force. Favouritism must fect. Eliminating them does not result in en­ be avoided. hanced motivation. 3.2.2.5 Rewards 3.2.4.1 Overtime Rewards can mean advancement in the Overtime generally means working in chain of command, social recognition, or mon­ excess of 40 hours a week. Most studies indi­ etary compensation. They may take the form cate that 40 hours a week is the optimum work of a pat on the shoulder or the satisfaction of a period and that working more hours reduces the job well done, depending on the circumstances rate of output. There are several reasons for and the character of the individual. But in all this slowdown. Workers tend to pace them­ cases, the worker should be aware of both the selves by slowing down to accommodate the reason and the nature of the reward. Moreover, longer day. The resulting productivity loss, ac­ the size of the reward should be commensurate cording to some sources, may exceed the time with the reason for it. Unearned or unduly worked beyond the normal 40-hour week. large rewards can have the opposite effect. Simply stated, after nine weeks of continuous Finally, rewards alone, without the other moti­ overtime, the output achieved in a 50-hour vating factors being satisfied, are of little value week is less than that which could have been on construction jobs. achieved in a 40-hour week. Table 3.1 illus­ n·ates the resulting loss in productivity with 3.2.3 Motivators overtime. Frederick Herzberg, one of the best­ This table has been derived principally known researchers of human behaviour, pro­ from a Detroit-area study performed in 1964. posed the following principles as the means of It fits quite well with other studies by the Me­ enhancing motivation: chanical Contractor's Association, and the • remove some controls while retaining ac­ Electrical Contractor's Association, a Proctor countability and Gamble evaluation and a major Engineer­ ing Procurement and Construction contractor's Table 3.1 Loss of productivity with overtime estimating guide. An alternative to overtime is altemate Inefficiency Factor work hours. Here are some examples of alter­ Days/Week Daily Weekly 7 Days 14 Days 21 Days 28 Days nate work hours: Hours Hours Four 10-hour days have lower daily start­ 5 9 45 1.03 1.05 1.07 1.1 up costs, reduced equipment downtime, levelled peak staffing demands, and de­ 5 10 50 1.06 1.08 1.12 1.14 creased absenteeism. 5 11 55 1.1 1.14 1.16 1.2 If a project has been satisfactorily complet­ 6 9 54 1.05 1.07 1.1 1.12 ed ahead of schedule, the construction 6 10 60 1.08 1.12 1.16 1.21 crew might receive some time off- with pay - for their extra effort. Or if workers 6 12 72 1.13 1.2 1.26 1.32 satisfactorily complete their assigned 7 8 56 1.1 1.15 1.2 1.25 amount of work, commensurate with an 8- 7 9 63 1.12 1.19 1.24 1.31 hour workday, they can go home, yet still receive a full day's pay. 7 10 70 1.15 1.23 1.3 1.38 7 12 84 1.21 1.32 1.42 1.53 With rolling fours, workers work 10- hour shifts: on 4 days, off 4 days. This type of

PRODUCTIVITY IN CONSTRUCTION ------~

work week reduces on-site population, decreases tivities is not coordinated. As a result, newly overall time for completion, reduces equipment completed work often has to be tom out. Such demands, and avoids fatigue by cycling different congestion can also give rise to unsafe practices groups of employees every 4 days. and conditions and leads to lost productivity in all trades involved. 3.2.4.2 Overstaffing Overstaffing occurs when more work­ 3.2.4.4 Crowding ers are assigned to a task than are required to Crowding can be considered in a man­ work productively. Overstaffing may take the ner similar to the scheduled acceleration of tasks form of increased crew size (for a given opera­ because the contractor attempts to complete tion) or the deployment of multiple crews; in more work activities in the same period of time either case, a loss of productivity will occur. or a designated amount of work in a shorter peri­ Figure 3.1 shows the effect of increasing crew od of time. More workers are placed in a given size over the number required to perform a task space than can function effectively. within the allocated time. Figure 3.2 illustrates the upper limit of the loss in efficiency with the percentage of Figure 3.1. Effect of crew overloading (overstaffing} crowding. The meaning of crowding is subject to a wide interpretation. Crowding occurs when (Optimum) 1 00 the work space per worker is reduced below a minimum required to work effectively. For ex­ 90 ample, if 18 workers are in an area that can only accommodate 15, the overcrowding is 3115 = 20%. According to Figure 3.2, 20% overcrowd­ ing results in an 8% efficiency loss, which is equivalent to an 8% increase in the normal dura­ tion of all activities being performed in the work 60 area during the period of overcrowding. (Figures 3.1 and 3.2 are meant to serve 20 40 60 80 100 only as a general guide; no precise information should be derived from them.) %Crew size increase (above optimum) 3.2.4.5 Multiple shifts

Introducing multiple shifts is another Adapted from: U.S. Optimum crew size for an activity rep­ less distractive way of adding more workers to Department of the Army resents a balance between an acceptable rate of the workforce. Double- or even triple- shifting Office of the Chief of En­ progress and the highest possible level of pro­ gineers. 1979. Modifica­ can be a reasonably economical method of ac­ ductivity. Experience shows that on a greatly tion Impact Evaluation complishing more work within the same period overstaffed project, the rate of progress may, at Guide. Washington, D.C. of time, but depending on the type of work, it 20314, p. 4-14. times, be improved by reducing the number of can also give rise to a chaotic situation. Trades workers or equipment on the site. Overstaffing requiring fine motor skills are ill-suited for dou­ dilutes supervision, slows down material deliv­ ble-shifting; where activities require high preci­ ery because of competing demands and, in gen­ sion, overall output may be even lower with a eral, affects the morale of the workers. double shift than it would have been with a sin­ The optimum crew size is the mini­ gle. Gross motor skill trades, on the other hand, mum number of workers required to economi­ and equipment operation, such as bulk excava­ cally complete a task within the scheduled time tion or building an earth-fill dam, can be double­ frame. As the number of workers is increased shifted very effectively. or decreased from optimal level, productivity A second shift, one that starts after the will vary proportionally. regular shift (i.e., after 5:00p.m.) is less produc­ 3.2.4.3 Stacking of trades tive than the regular shift. People who work shifts face many problems that other workers do Stacking of trades (creating conges­ not. These problems come from changing eating, tion) is a problem that develops when different sleeping, and working patterns. trades, which should be working sequentially, are obliged to work simultaneously in a limited When shift cycles are changed, the first work space. When this occurs, the work area several days are periods of change and employ­ becomes smaller (or at least, appears so) be­ ees will be less alert, less accurate, and less safe. cause all trades are trying to bring in the mater­ Sometimes shift rotation is invoked as a means ial required for their work. Each trade tries to to be fair to all workers, but it is actually unfair. complete its work but the sequence of their ac- It takes almost a month for the human body to

PRODUCTIVITY IN CONSTRUCTION 1. inadequate tools and equipment Figure 3.2 Effect of congestion of trades (crowding) 2. excessive owner surveys of on-site work 3. poor planning

>. 16 4. poor overall management () :; c 0 -~ 5. mediocre supervision .0 () 12 -:t::(1j ·- Q) c c 6. overtime available on another job site Q) ·- 8 ~.8 Q) rn 7. unsatisfactory relationship with boss. D... rn .Q 4 Many of the reasons for absenteeeism and turnover can be affected by management. 0~~~----~--~----~--~----~--~ 0 5 10 15 20 25 30 35 By simply being aware of their major causes, supervisors may be able to make improvements %Crowding on their sites.

Adapted from: U.S. Absenteeism and turnover have the following negative effects on productivity: Department of the Army adjust to a different schedule. Moving workers Office of the Chief of En­ • Crew members waste time waiting for re­ back and forth from shift to shift does not let gineers. 1979. Modifica­ placements. them adjust to a schedule and consequently tion Impact Evaluation • Time is spent transporting replacements to Guide. Washington, D.C. they will not perform at their best. 20314, p. 4-14. and from other work locations. 3.2.4.6 Stop-and-go operation • Supervisors lose time in reassigning work activities and in locating replacements. Stop-and-go operation occurs when an essential component of an activity is not avail­ Other losses are incurred from not able when it is required. The component might having the workers available, administrative be a drawing, a decision about a contemplated costs (payroll, personnel, etc.) for terminating change, the acceptability of workmanship, pre­ and hiring people and the disruption to fellow purchased material, or equipment. The activity workers. is halted temporarily and the crew moved else­ On average, it is estimated that 24 p­ where to a new task. Breaking the rhythm, tak­ hs of paid time are wasted for each resignation. ing time to make a decision on the next step (usually referred to as reaction time), packing 3.3 Human Factors Related up tools, moving to the next activity, unpack­ ing, orientation, and obtaining the required sup­ to Productivity plies, are all non-productive activities. Human factors related to productivity Additional labour input is required without a fall into two groups: corresponding increase in output, resulting in a • Individual factors, such as personal attrib­ net loss of productivity. At times, losses can be utes, physical limitations, the learning in the order of 30 to 40%. curve, teamwork and motivation; 3.2.5 Absenteeism and turnover • The worker's environment, such as cli­ mate, work space, and noise. The major reasons for absenteeism, listed here in order of importance in the con­ Since construction work is labour-in­ struction industry, are: tensive, site workers clearly play a major role in the construction process. Although human 1. personal or family illnesses factors are often not given much consideration, they strongly influence job site productivity 2. poor overall management and are key to the success of any project. 3. poor supervision 3.3.1 The individual as a factor 4. excessive travel distance to the job site affecting productivity 5. excessive rework Persons with an optimistic and posi­ 6. unsafe working conditions. tive attitude are likely to have more initiative and think of imaginative solutions to various The major reasons for turnover in the problems. A caring, considerate, and friendly construction industry, also listed in order of im­ person with a sense of humour can help in­ portance, are: crease productivity. Humour in the workplace puts people in good spirits, relieves stress, and develops teamwork.

PRODUCTIVITY IN CONSTRUCTION A safe and healthy person is more pro­ continue working for long periods of time. ductive. Respect for safety and safe practices However, if the work requires more than 17 kJ must be encouraged, not only for the well­ (4 kcal.) of energy per minute, the reservoir being of the workers, but to minimize 'down­ will drain and when it empties, they require rest time' on a project. to refill it with energy. A creatively thinking person can con­ "For an average construction task requiring tribute to increased productivity. Often it is the 6 kilocalories [25 kJ] per minute including workers who come up with the best solution to a basic metabolism, work at this pace could problem. Workers who demonstrate leadership continue for no longer than 25 minutes be­ skills should be encouraged to develop their po­ fore the worker becomes exhausted. An av­ tential because construction crews need good erage male, sawing and hammering with an leaders to be successful and productive. Leader­ energy demand of 8.1 kilocalories [34 kJ] ship skills include such characteristics as hon­ per minute, must rest after about 8 minutes." esty, responsibility, good judgment, co-operation, (Ogelsby et al., 1989) being organized, and being a good listener. To avoid short-term fatigue, tasks Finally, experience plays an important should be designed to avoid activities such as role in the productivity of a worker. holding heavy loads or pushing hard against non-moving objects. Use tables, supports, 3.3.2 Physical limitations props, jigs, fixtures and tools or other devices Humans are somewhat like machines as a substitute for muscular effort. in the sense that they require fuel to operate and The right amount and type of tools can increase produce energy (the capacity to do work), and productivity. Cutting and welding torches and they become exhausted if they are not looked welding-rod holders should be positioned to re­ after properly. Many construction tasks are duce effort and make work more visible. physically demanding. Sanders, grinders, drills, hacksaws and similar The type of work that persons are per­ tools should have good weight balance and forming will dictate how frequently they need handgrips. Wheelbarrows and buggies should to rest and regain energy to continue working. be designed so weights are balanced, thus re­ Figure 3.3 illustrates this with a water reservoir quiring little lifting. Pneumatic tires increase analogy. An average young male adult can de­ ease in pushing and guiding. velop approximately 21 kJ (5 kcal.) of energy If a worker has to put himself in an per minute, of which approximately 4.18 kJ (1 awkward position to perform a particular task, kcal.) per minute is needed to sustain life and it can lead to discomfort and even injury. Per­ the rest is available for expenditure in the form sons working in an uncomfortable position are of work. If workers perform light work, then more likely to take breaks and work less pro­ the energy reservoir remains full and they can ductively. Working overhead tires the arms and can put the back in odd positions. Constant bending also puts unnecessary strain on the Figure 3.3 Water-tank analogy of the human body's back. Back injuries are very common in the energy storage-replenishment capacity (Ogelsby et construction industry and these could be avoid­ al., 1989} ed if more work were done at waist-height. Male 25 years old, good physical condition 3.3.3 The learning curve Max. input 21 kJ/min The first time any person performs a certain task, they will work slowly because they are learning how to do it. With additional repetitions, the time needed to perform the Reservoir same or similar tasks will decrease. It is there­ Capacity fore desirable, where possible, to have the same 10.5 kJ person perform a task several times rather than making personnel changes along the way. After a considerable number of repetitions, the learning curve approaches a plateau that re­ Light to medium work Basal metabolism flects the minimum time required to perform a < 17 kJ/min (life sustaining) task (Figure 3.4). 4.18 kJ/min Heavy work> 17 kJ/min This principle applies to highly repeti­ Reservoir draws down tive manual operations. If delays occur be­ tween repetitions, the 'unlearning curve' effect can be noted as the worker gets out of practice Recovery rate= 21 kJ/min minus 6.3 kJ/min for rest equals 14.6 kJ/min and can no longer perform the task as well. It

PRODUCTIVITY IN CONSTRUCTION ------People enjoy not only the challenge of meeting Figure 3.4 The learning curve and exceeding production targets but also con­

100 ~.,r--,--~--~---.--,---,---~--~-, tributing to solutions to problems. Mild com­ petition in production objectives is also healthy and useful, i.e., productivity competitions be­ 90 tween crews or between shifts. Supervisors can achieve higher levels of productivity by appeal­ 80 ing to a worker's pride, competence, sense of duty, and team play. 70 3.3.5 Environmental factors "Other things being equal, human be­ 60 ings perform relatively continuous physical or mental work most effectively when the temper­

50 L---~--~--~--~~~~~~--~--~--~ ature falls between 10 and 21 'Cat a relative hu­ 0 2 4 6 8 10 12 14 16 18 20 midity (R.H.) of 30 to 80%, under dry conditions, with the atmosphere clear of dust Cumulative Units and other atmospheric pollutants, and without excessive noise. Departures from these condi­ tions have adverse effects on productivity, com­ takes time for the worker to re-learn how to do fort, safety and health" (Ogelsby et al., 1989). the task. The same effect will be noted after personnel changes are made as the new workers 3.3.5.1 Weather conditions must learn what to do. The unlearning curve is Workers must slowly become accli­ illustrated in Figure 3.5. matized to working in hot weather. Heat stress 3.3.4 Crews and teamwork occurs at temperatures above 49·c (120.F) at an R.H. of 10% and 31 ·c (88'F) at an R.H. of Construction usually requires that a 100%. Above these temperatures, heat injuries group of diverse workers act as a team with can occur, which include sunburn, heat cramps, specific objectives. Teamwork can be main- heat exhaustion, and heat stroke. These illness­ es can be prevented by using acclimatization, Figure 3.5 'Unlearning' curve adequate rest periods, proper clothing, and ade­ quate water and salt intake. Similarly, the ill effects of cold weath­ er can be warded off by wearing proper clothing 90 and having temporary shelters near the work area; heaters may be installed as long as they are well ventilated. The optimal temperature 80 appears to be S"C. At this temperature the pro­ ductivity of indoor work is not greatly affected. 70 Table 3.2 shows the reduction of work efficiency in cold weather. It is assumed that 60 Interruption efficiency is 100% at 21 ·c (70.F). time

50 L-~--~------~--~--~--._--~--~~ Table 3.2 Reduction in work effi· 0 2 4 4 6 8 10 12 14 16 ciency in cold weather Cumulative Units Loss In Efficiency(%) tained or improved by good, open, two-way Temp. ·c Gross Fine communication. In that vein, workers should Skills Skills be asked for suggestions and solutions. Not 4 0 15 only does this make workers feel that their -2 0 20 opinions are valued and important, but it usual­ ly results in a solution to the problem. -7 0 35 -13 5 50 This idea was developed in Japan through the use of quality circles. Groups of -18 10 60 workers would meet and develop solutions to -23 20 80 problems in their work, which would then be -28 25 90-95+ {probably presented for management action. Supervisors can't work) and managers should aim to produce a produc­ -34 35 tive environment and set goals for the team.

PRODUCTIVITY IN CONSTRUCTION ------the ill effects of the environment on the work­ Table 3.3 Relationship of temperature and humidity ers' productivity are minimized. to productivity 3.4 Job Site Planning R.H. Temperature ('C) -23 -18 -12 -7 -1 4 10 16 21 27 32 38 43 Planning a site for the most efficient 90 56 71 82 89 93 96 98 98 96 93 84 57 0 use of all construction facilities leads to im­ proved productivity for everyone. The degree 80 57 73 84 91 95 98 100 100 98 95 87 68 15 of planning will depend on the complexity and 70 59 75 86 93 97 99 100 100 99 97 90 76 50 size of the job. A good job site plan, the result 60 60 76 87 94 98 100 100 100 100 98 93 80 57 of good job site planning, is of utmost impor­ tance to ensuring a productive workplace, re­ 50 61 77 88 94 98 100 100 100 100 99 94 82 60 gardless of project size. 40 62 78 88 94 98 100 100 100 100 99 94 84 63 30 62 78 88 94 98 100 100 100 100 99 93 83 62 3.4.1 Job site planning considera· tions 20 62 78 88 94 98 100 100 100 100 99 93 82 61 The preparation of one or more draw­ ings with accompanying text should form the Table 3.3 shows the combined effect basis of the job site plan. A documented plan is of temperature and relative humidity on pro­ equivalent to building the project on paper, ductivity. This table, developed by the Nation­ where mistakes can easily be corrected and al­ al Electrical Contractors' Association, can be ternatives can be tested at little cost. used to predict the effects of weather on pro­ ductivity for most construction tasks. This in­ Many government agencies, utilities, formation can be useful when planning and and traffic planners require the information in­ estimating work and preparing construction dicated on a site plan before issuing various claims. construction-related permits. Timely receipt of these permits is required in order to avoid de­ 3.3.5.2 Noise lays. Completion of the plan with subsequent revisions will require input from the entire pro­ Noise can interfere with work. It may ject team. create a safety hazard by not letting workers hear warnings or instructions. Noise may not In conjunction with the key trades, a affect the amount of work accomplished, but it plan for excavation, shoring, and de-watering will affect the quality, especially if concentra­ should be developed. In the development of tion is required. Moderate, steady background this plan, all previously established information noise, such as music, may actually increase per­ from the preliminary site plan, with particular formance. It covers up random, disruptive consideration to adjacent property encroach­ sounds and sets a pace to work by. Studies ments and living restrictions, is used. suggest that 90 decibels is the noise level at which possible hearing damage and decreased Considerable lead time is required to work performance result. address environmental concerns properly. A plan has to be established at the project plan­ Effects of noise can be mitigated by ning stage for the removal of contaminated ma­ reducing the noise at the source, separating terials and other substances, if such work has noise sources from the workers, or having been identified. workers wear ear-plugs or other protective gear. Temporary access points for initial and final excavation ramps should be estab­ 3.3.6 Workspace lished to minimize interference with temporary construction services and be coordinated with The workspace should be set up so the actual succeeding construction program, that it provides workers with a safe, healthy, permanent hoarding entrances, and traffic flow and comfortable environment. It should be or­ requirements. ganized in an efficient and appropriate manner for the nature of the work being done. 3.4.2 Temporary electrical service Spending the time to keep the site The power requirements and the avail­ clean is worth the effort, because it helps keep ability of power on a project should be re­ the project organized. Workers who feel that viewed as soon as possible to avoid the worksite is safe will be more productive. unnecessary delays once the project has started. The workspace should also be well lit Furthermore, to avoid delays on the and well ventilated, and comfortable, so that project associated with inadequate electrical

PRODUCTIVITY IN CONSTRUCTION ., ------distribution, this activity must be planned in de­ sanitary facilities should be placed in reason­ tail by individuals knowledgeable in the field. able locations to ensure that the workers do not Too often trades are faced with electrical distri­ spend unnecessary unproductive time travelling bution systems that are inadequately sized, to and from their work station. lacking in sufficient outlets, or unsafe to use. The inability to operate electrical tools and 3.5 Safety Issues equipment effectively will result in an obvious Everyone, from owners to workers, loss in productivity. Associated with this prob­ benefits from a safe construction environment. lem is that of an improper temporary lighting In the context of macro-productivity, a safe site system which can produce safety concerns, an is a productive site. The Business Roundtable unproductive work environment and potential Study, "Improving Construction Safety Perfor­ labour disruptions. mance," concluded that accidents cost $8.9 bil­ 3.4.3 Temporary heating and lion (U.S.) or 6.5% of the $137 billion spent on hoarding industrial, commercial, and utility construction in the United States in 1979. This 6.5% figure The type of hoarding, the amount of is probably low. If the same percentage is ap­ heat required and the time for which it has to be plied to all types of construction in Canada, applied, depend upon the work occurring at the then the cost of accidents today may be conser­ particular area or stage of the project during the vatively estimated at over $5 billion. winter months. There are entirely different re­ quirements for rough carpentry than for taping The apparent high cost of accidents in or painting. All of these considerations have to construction easily justifies expenditures on be taken into account when establishing the construction safety. While owners, construc­ heating system. tion managers, and contractors have long rec­ ognized a moral obligation to provide a safe As with the temporary electrical ser­ work environment, the economic reasons may vice, the gas utility companies require suffi­ not have appeared so compelling. cient lead time in which to provide temporary services. 3.5.1 Economic impact of acci· dents An undersized heating system, a sys­ tem that does not protect the entire work area, Three basic cost categories are directly or an inadequate hoarding system will ultimate­ related to accidents. These costs include com­ ly produce downtime on a project. Choose pensation for injured workers, liability claims, high-capacity heaters with ducted fresh-air sup­ and property losses. ply, where possible, to provide cleaner, dryer Direct compensation costs for injured heat, and better fuel efficiency. workers are largely made up of the cost of com­ 3.4.4 Miscellaneous systems pulsory injured workers' compensation insur­ ance. In Ontario, for example, a contractor Job site efficiency can usually be im­ involved in steel erection was required to pay proved by using: $25.67 for every $100 of wages paid to an iron­ • communications systems as part of a prop­ worker in 1988. Such compulsory insurance erly organized project payments are based upon the loss or claim his­ • a central compressed-air system accessible tory of the particular class of construction by all trades workers. The median worker's compensation • a water distribution system costs were found to be approximately 1.9% of • a refuse removal system. the total project costs; these costs vary from 1% to 4% of the total project costs. The cost of supplying these systems is minimal compared to the time lost by not hav­ While the cost of property loss may be ing them. quite small compared to workers' compensa­ tion costs, the resulting indirect losses due to 3.4.6 Offices, lunchrooms, and damaged property may be substantial when ex­ sanitary facilities pressed in terms of a ratio of direct costs. In The location of owner, consultant, one study (The Business Roundtable 1982), the general contractor, and sub-trade offices should ratio of indirect to direct costs was found to be be as close to the site as possible and close to approximately 5: I for various cost categories. each other. Each site has its own characteris­ 3.5.2 Safety and productivity tics and requirements for office locations. Sim­ ilarly, the job site plan should carefully Most construction accidents occur consider the location of the on-site workers' during non-productive periods. Sloppy job storage and lunch facilities, keeping in mind sites reduce productivity and increase the local union requirements, if applicable. Good chances for accidents to occur. Management

PRODUCTIVITY IN CONSTRUCTION ------., should play an active role in ensuring safety. of the project. Certainly the effects of weather, Craftsmen are more productive when they safety, and site congestion must be accounted know that management is genuinely concerned for during estimating. This chapter has provid­ about their well-being. ed an overview of these human resource issues and their relation to productivity. Workers who are more likely to have an accident are those with a bad attitude, those Additional Readings who are frequently absent- especially on Mon­ days and Fridays (Hinze 1981)- and those who Ahuja, H. A. 1984. Project Management Tech­ have a history of accidents. Pre-employment niques in Planning and Controlling Construc­ screening keeps workers with poor records or tion Projects. New York: John Wiley and high accident potential off the job or at least out Sons. of hazardous situations. Orientation, which in­ cludes training, and attention to new workers, is Carlson, J. G. 1961. "How Management Can especially important. Twenty-four percent of Use the Improvement Phenomenon." Calffor­ all accidents happen to workers in the first nia Management Review 3 (2):83-94. month of work and 46% occur in the first six Gates, M. and A. Scarpa. 1972. "Learning and months. Experience Curves." ASCE J. Canst. Div. 98 All workers new to the crew should be (C02):79-101. assessed by asking about their previous work Gates, M. and A. Scarpa. 1987. "Optimum experience, and closely supervised. If new Number of Crews." ASCE J. Canst. Div. 104 workers are to work as part of a group, the su­ (C02): 123-132. pervisor must ensure that they are accepted by the group. Those who are to work alone, Halpin, D.W. and R.W. Woodhead. 1976. De­ should be put to work only after their skills sign of Construction and Process Operations. have been fully tested and the supervisor has New York: John Wiley and Sons. made sure that all procedures are understood. Hendrickson, C. and T. Au. 1989. Project Instruction and toolbox meetings Management for Construction, Englewood should be relevant to ongoing work. Assign­ Cliffs, NJ: Prentice-Hall. ments and their safety implications should be discussed and meetings held when safety re­ Hinze, J. 1981. "Biorhythm Cycles and Injury quirements change. These occasions can be Occurrences." ASCE J. Canst. Div. used to stress safety and that unsafe work prac­ 107(1):p.21. tices will not be tolerated. Oglesby C.H., H.W. Parker, and G.A. Rowel. Pressuring the crew or individuals if 1989. Productivity Improvement in Construc­ there is a productivity problem may cause tion. New York: McGraw Hill. workers to work more quickly and less safely. The Business Roundtable. 1982. Report A-3. Instead, productivity problems and solutions The Business Roundtable, 200 Park Ave., New should be discussed. Safety is good business: it York, N.Y. affects worker morale and attitude, and has an economic impact on the project. Thomas, R., C. Mathews, and J. Ward. 1986. "Learning Curve Models of Construction." J. The human resource is extremely im­ Canst. Eng. and Management 112 (2):245-257. portant in construction, more so than in any other industry. This is simply because con­ Touran, A., A. Burkhart, and Z. Qabbani. 1988. struction projects are unique and complex. "Learning Curve Application in Formwork These characteristics inhibit full automation Construction." In Proc. of 24th ASCE Annual compared to other industries. The individual Conference, 20-24. San Luis, California. skill of each craftsman, the abilities to commu­ nicate, make decisions, work with others, and United Nations Economic Commission for Eu­ share information, makes this resource unique rope. 1965. Effect of Repetition on Building and irreplaceable in the foreseeable future. To Operations and Processes On Site. get the most out of this resource, the manager ST/ECE/HOU/14. has to realize what motivates the worker, what demotivates, what the physical limitations are, and what factors inhibit performance.

The understanding of these issues is important not only during the progress of con­ struction. At the bidding or planning stage, the estimator often has to make personal judgments about productivity under anticipated conditions

PRODUCTIVITY IN CONSTRUCTION ------.a 4 Measuring Productivity from the Cost-Reporting System

4.1 Introduction Productivity and productivity factors can be reviewed at any level required. Com­ On any project there are numerous ac­ puterization can produce overwhelming tivities that could be tracked. The most signifi­ amounts of data, much of it superfluous. The cant are usually those of greatest scheduled more data processed by a system, the costlier duration. In selecting the detail and extent of the process. A costing system can produce pro­ control of activities, the method should be kept ductivity data for a small additional cost. It is simple and only the degree of control needed these productivity data that are necessary to should be exercised. Therefore activities where gauge performance on the project. the maximum concentration of hours occurs should be the focus. 4.2 Data Collection and Processing Figure 4.1 Time card This section provides a simplified de­

TIME SHEET scription of the origin, collection and process­ ing of individual pieces of data for measuring Employee Name: ------Job No· productivity and work progress. Only the gen­ Employee No: Pay Period No: _ ___ Pay Period Date: ---- eral aspects of cost reporting are discussed. Date Reg O.T. T.T. Date Reg O.T. T. T More detailed discussion can be found in nu­ w merous books (for example, Halpin, 1985; T Adrian, 1979). F Data in construction projects are col­ s lected on various forms and for different pur­ s poses. Data collected for the financial control M system are organized primarily as required for T tax and other legal purposes. These data are not Total Total sufficient to control the cost of a project; addi­ Certified Correct Supervisor: ______tional information must be collected. To illus­ Employee: ------Remarks: ------trate: a financial system keeps track of the payroll of all workers on a project, but it does not necessarily account for the hours of labour spent on a particular work package. (A work Figure 4.2 Sample daily work report package is a group of related tasks.) For cost control and productivity calculations, the per­ Daily Work Report son-hours spent on the activity must be collect­ ed. Accordingly, a company normally Project: Prepared by: maintains a dual-purpose system. For financial Date: Comments: purposes, data are gathered at an aggregated Temperature: level; for cost control and productivity measure­ ment, data are tracked at a more detailed level. Weather conditions: Work Description Labour Supervision Craft-1 Craft-2 Craft-3 Total Data useful for cost control and pro­ pkge. (h) (h) (h) (h) (h) (h) ductivity measurement on a construction pro­ code ject are gathered in three categories: labour, 2-20 Concrete 64 8 - - - 72 material, and equipment. Variations in these formwork categories are tailored to suit the requirements of each company.

The data-collection process is initiated at the construction site by collecting the labour, equipment and material information from time cards (Figure 4.1), daily reports (Figure 4.2), and material-issue tickets. To make use of this

PRODUCTIVITY IN CONSTRUCTION Figure 4.3 Information flow summary on a typical Figure 4.5 Weekly material report reporting system Weekly Material Report Cost Control System Setup Supplier P.O.# Description Terms Cost Code Amount Big 2104 Forms Net 10 301 $4,002 Lumber Inc. Ready 2361 Concrete Net 10 305 $6,700 Mix Co.

3. Job Cost Ledger (301) 4. Job Cost Ledger (306) Concrete Placement Figure 4.6 Weekly equipment time sheet

Equip, Dally Cost Hours Amount • Rate Code c::>= '···-...•• ' - information it must be further refined and con­ [ WEEKLY EQUIPM~NT REPORT Coil Description Equipment Hours Rate Total solidated. The data are normally organized in Code what is referred to as a cost-reporting and con­ 234 Excav at 12T. Truck 30 60 1800 Sile B trol system, either computerized or manual. A simplified cost-reporting system is illustrated _ in the flow chart displayed in Figure 4.3.

The most basic information in this The data from the first source flows into system is obtained from three sources: the weekly labour, equipment, and material re­ o labour time cards, daily reports and pay­ ports (Figures 4.4 through 4.6). The weekly re­ roll records ports summarize all information by work o weekly field quantity reports package (with unique cost codes). The informa­ o the original estimate (budget). tion collected represents the labour hours spent, rates, and total labour cost on each work pack­ age. The accuracy of the entire system de­ Figure 4.4 Weekly labour report pends on the correct application of the hours

0~~ Labour Time Card worked to their respective work packages, iden­ Crall Cos I Tolal Employee Amaunl Aale Code Hours tified with cost codes. The weekly labour, equipment, and material time sheets are subse­ c::>= c;::>= quently used to compile a weekly distribution report for labour, equipment, and material, as ! WEEKLY LABOUR REPORT illustrated in Figure 4.3. The weekly distribu­ Cost Regular Overtime Regular Overtime Total Code Hours Hours Rate Rate tion report summarizes, for every work pack­ 301 ® ® 14 21 490 age, the labour expenditure for the given week and the quantities placed in that time. The 342 8 \ 4 \ 14 21 84 ' quantities are obtained from the field quantity \ report shown in Figure 4.7. : \

Total hours spent on package 301 is 30 person-hours lor this period This is translerred to job cost ledger lo measure productivity on this package.

PRODUCTIVITY IN CONSTRUCTION ______.,

The weekly labour distribution report Figure 4. 7 Weekly field quantity shown in Figure 4.8 becomes the basis for fur­ report ther data reductions, as illustrated by the infor­ mation flow shown in Figure 4.3. Weekly Field Quantity Report Work Completed This Week: The weekly distribution reports on labour, equipment, and material are used to Cost Description Unit Total Total compile two additional reports, namely the job Code Completed to This Period Date cost ledgers and the job cost journal. The job cost ledger is prepared for every work package. 302 Concrete Placed C.M. The job cost ledger (Figure 4.9) summarizes all 304 C.M. expenditures on the given task including labour, equipment, material, and subcontracts. The ledger also shows the total expenditure on the task to date, the estimated quantity of work, estimated total cost, total person-hours spent to date, and the productivity achieved on this task. Figure 4.8 Weekly labour distribution report For each cost ledger, productivity rates in cost/unit and person-hours/unit can be WEEKLY LABOUR DISTRIBUTION REPORT calculated. The entries in the job cost ledger From show an example of the task of concrete plac­ From Weekly From ing. The total cost to date (sum of Col. 6) is Weekly Labour Quantities Estimates Report $26 510. The quantity excavated to date from ! the field quantity report is 1000 cubic metres / ' (C.M.). The weekly labour report shows that Cost Labour Cost \ Quantities Estimated Cost Description \ the total number of p-hs spent on the task to Code • Week Cumulative , Week Cumulative Per L!nit Total ·. date is 102. The productivity is estimated as 302 Concrete Fmwk 1240 5020 400 1200 $2.5 $7500 $26 510/1000 C.M. or $2.65/C.M. and 102 per­ son-hours/1000 C.M. (or 0.10 p-h/C.M.). The 1- original estimate shows that the unit cost on :;:; 2 this work package is $2.45/C.M. and the p-h budgeted per unit is 0.15 person-hour/C.M. A ""' ~ ) Labour "' comparison of actual versus estimated reveals Total/Expenditure: This period ; $6200 V negative cost and positive person-hour vari­ To date ; $15400 ances. The cause of the variances could be the hourly labour cost and better unit production rates or other factors, such as better than antici­ pated equipment. The job cost ledgers are summarized Figure 4.9 Job cost ledger into one document as the cost ledger summary (Figure 4.10). This document provides an ef­ JOB COST LEDGER fective means of viewing the status of the pro­ Cost Center Name: Concrete Placing Code 301 ject by displaying a summary of each task. Source Description Labour Material Equipment Subcontracts Other Total The job cost journal (Figure 4.11) is a LD-12 Labour Distribution 341 341 weekly itemization of all expenditures. It Report #12 shows total costs for material, labour, equip­ C-123 ACE Hardware 73 73 ment, etc., for the entire project. Another use LD-1 237 237 y;yz Excv. Inc 1800 1800 of this document is to double-check the work 578 73 1800 2451 done to date. (The totals on the job cost journal 578 73 1800 2451 and the job ledger summary should match.) 578 73 1800 2451 200 200 Productivity, as defined earlier, is 200 200 input/output, i.e., person-hours used to install an amount of mate1ial. This is shown at the bottom of the job cost ledger (Figure 4.9). Totals ------~ The final pieces of the job costing sys­ Estimated Quantity; __ tem, as illustrated in Figure 4.3, are a labour Tomls From Woefdv L.obour flcpOij Productivity: • Person-Hours ---- - fOI ""'"""'Ill& 102 P·ll cost performance report and a project monthly Ouanuty ------from Reid Quantity Reports progress report. The labour performance report ...- 0. t P·HIC.M ,--- - For our IWimPle:. 1000 C M 1 . ' (Figure 4.12) draws information from the esti­ Total Cost$ --- _..~._ ------·------= Quantity -- -- - ! mate and the weekly distribution report for = $2.65/C M labour, reduces it, and presents it by task. For

PRODUCTIVITY IN CONSTRUCTION tions for this period and the accumulated costs Figure 4.10 Cost ledger summary to date. The total columns present the forecast and the estimated total costs for each. In the far COST LEDGER SUMMARY right-hand column, the 'percent complete' Cost Description Quantity Labour Material Subcontracts Other Total based on quantity installed and cost incurred is Code shown. This is a complete summary of the sta­ 302 tus of the project labour cost as of the reporting 304 [ Summary of each cost account from the Job Cost Ledgers period. This information enables management I to ascertain the status of each package and how well the whole project is progressing. Achieved productivity in cost/unit for all work packages are displayed next to the estimated cost/unit Totals should agree with Job Cost Journal rates. The manager is then able to view all ( tasks, compare their status to the original esti­ I I mates, and take corrective action where needed. The final report is the project monthly report, which is mainly compiled from the job Figure 4.11 Job cost journal cost ledgers and the original estimates. The in­ formation for this report is shown in Figure 4.13. JOB COST JOURNAL The illustration contains five sets of information: # Cost Period Description Labour Material Subcontract Other Total • actual quantities, costs and unit costs Code • quantities and costs required to complete 1 Wk1 Payroll 3004 3004 the package 2 Wk 1 Payroll taxes 290 290 • a forecast of the total cost 3 Wk1 P.O.# __ 5000 300 5300 • the estimated quantities, unit costs, and Freight __ total costs 4 • variance from estimate, as well as an idea of performance. [ From Labour Distribution Report J This final monthly report, derived [ From Weekly Material Report ] from the cost-reporting system, summarizes construction progress and the productivity for Total Total Total Total Labour Material Subcontract Others Total the month. 4.3 Tracking Person-hours instead of Costs in the Cost-reporting System Figure 4.12 Labour performance report The basis of cost control is the dollar LABOUR PERFORMANCE REPORT spent on a given task. Productivity is also mea­ Estimated Cos I Quantities Cost Per Unit Variance Total Cost sured as cost/unit. Although the measure of Code Description l--.----.-+---.--r--l--- -.-- -l---.,...--1---%-co_m-rple_le-l person-hours/unit is possible for each package, Per. Cum . Est. Per. Cum Est. Period Cum , Forecast Est Quantity Cost it is not the basis of the system. It can instead 302 Concrete 300 600 3300 0.2 0.25 0.3 30 ~0 825 990 18.2 15.2 lormwork ~ • : ~' track p-h expenditure, rather than dollar values, 304 :: ..·· in the control system. Both techniques have merit. Advanced computer tools provide sys­ tems that can easily report both. Three mea­ sures of output - estimated 'percent complete,' physical measurement, and earned value - can : be derived from a person-hour-based system.

(0,2 · 0.3) X 300 4.3.1 Estimated 'percent complete' = 30 The estimated 'percent complete' is simple and a relatively inexpensive measure each task, the quantities installed during the re­ used to calculate the quantity completed. The porting period (one week in this case), the cu­ calculation is: mulative installed to date for the package, and Estimated Quantity Completed = Total the original estimated quantity are presented. Quantity x Estimated Percent Complete Under the heading, cost ($) per unit, the unit This measure has two disadvantages: costs for the period, cumulative unit costs to • Estimated 'percent complete' is subjective date, and the unit costs in the bid estimate are (someone's guess). shown. The variance columns indicate positive • This method is not sensitive to changes in (underrun) and negative (overrun) cost devia- scope of work.

PRODUCTIVITY IN CONSTRUCTION Figure 4. f 3 Monthly cost summary

MONTHLY COST SUMMARY Actuals To Complete Forecast Cost Estimated Variance Index Cost Description Actual Actual Unit Quantity Cost Code Quantity Cost Cost to to to date to date to date Complete Complete Quantity Unit Cost Total (AQ) (~C) (AUC) (CQ) (CC) (FC) (EQ) (EUC) (EC) . . AC/AQ EQ-AQ CQ x AUC ACt CC EC- FC EC I FC .. . ' From Job Cost Ledger for this [ account / I

Accordingly the 'percent complete' milestones or partial completion. In form work, method is used for such straightforward items for example, it can be agreed upon, in advance as masonry. of the work being undertaken, that credit will be taken as follows: 4.3.2 Physical measurement • fabricate 60% The physical measurement requires • erect 20% the actual counting or measuring of the number • remove 15% of work units completed. Examples of work • clean forms 5%. units are diameter centimetres of pipe welds, Table 4.1 shows a typical performance number of doors hung, and square metres of report. Column 11 shows 645 p-hs earned formwork installed. This method is objective based on the rules of credit. and detailed, and scope changes are accurately included. Physical measurement is, however, 4.3.4 Performance factors time-consuming and expensive, and its use is Of two approaches for measuring pro­ generally restricted to tracking bulk material ductivity on sites, the most common one is quantities, especially in fabrication shops. based on the use of performance factors (PFs) 4.3.3 Earned value that can be determined from data produced by the costing system and the EV measure. By de­ The earned value (EV), a measure finition, widely used in construction, is a technique for PF = Earned p-hs/Actual p-hs calculating the 'percent complete' of a control (Cols. 13 and 14 in Table 4.1) account. It is more objective than the estimated 'percent complete' measure but not as detailed Earned p-hs are calculated based on and expensive as the physical measurement. estimated values for unit p-h, e.g., p-h/tonne. EV is, in a sense, a compromise between the The use of actual p-h (Cols. 1 and 2 in Table two measures. The p-h input is taken from the 4.1) and estimated values of unit p-h (Col. 7 in craftsmen's time cards. Only direct work is Table 4.1) can lead to inaccuracies because the used for calculating productivity. estimate could be wrong, hours worked could have been charged incorrectly to accounts, and For each code of accounts, the fore­ measurement of the physical progress could be man reports the actual quantities installed. inaccurate. Based on some rules for taking credit or on the estimated 'percent complete,' credit is taken PFs are used to help in the control of a and an earned value is calculated as follows: project. Figure 4.14 graphically shows cumula­ tive and time period productivity. Performance EV =actual quantities x estimated (or regarding the project or a particular account are budgeted) productivity per unit of quantity out­ readily evident. Besides the actual value of PF, put. it is very important to interpret the trends dis­ played by these curves. Initial productivity can For example, p-hs earned= actual be expected to be low (0.9) because the start-up C.M. installed x estimated p-h/C.M. activities are, on average, more time-consum­ Rules of credit provide a structured ing than other activities. With time, repetition method of allowing credit for intermediate and familiarity contribute to lower p-h rates per unit and improved productivity.

PRODUCTIVITY IN CONSTRUCTION ------Table 4.1 Typical person-hour performance report

Activity Acct. Actual. Quantities Unit Budget Earned Performance Projected No. p-hs p-hs p-hs p-hs Factor p-hs This To Current This To Unit Budget This To This To This To Period Date Budget Period Date Meas. Period Date Period Date Period Date (1) (2) (3) (4) (5) (6) (7) (B) (9) (10) (11) (12) (13) (14) (15) Formwork 3.03100 680 37,100 82,000 620 34,000 m2 1.04 1.10 1.09 85,200 645 36,380 0.95 0.98 86,940

Notes Column 1 Actual p-hs for latest reporting period [from daily time sheets (see Figure 4.4)] 2 Summation of actual p-hs 3 Total estimated quantity for the account 4 Earned quantity in place for the latest reporting period [from quantity in place report (Figure 4. 7)] 5 Summation of earned quantities in place 6 Unit of measurement 7 Budgeted productivity rate (based upon historical records) 8 Col. 8 = Col. 1 + Col. 4 9 Col. 9 = Col. 2 + Col. 5 10 Col. 10 = Col. 3 + Col. 7 11 Col. 11 = Col. 4 +Col. 7 12 Col. 12 = Col. 5 + Col. 7 13 Col. 13 = Col. 11 + Col. 1 14 Col. 14 =Col. 12 +Col. 2 15 Col. 15 =Col. 10 +Col. 14

4.4 Cost Reporting and Figure 4.14 Trends of productivity factors Analysis Using Project 1.20 ,-----,,-----r----r--.--.---.--...,--.--.---, Management Software 1.15 Forecast PF The introduction in the preceding sec­ at completion tions to a manual cost-reporting system con­ 1.10 veys an understanding of the basic concepts: 0 how data are reported and accumulated, how to t3 t1l 1.05 u. feed information into the various computer pro­ grams, and how to interpret the output. ·:;:.?!' 1.00 n To illustrate how a commercial pack­ :::J 0.95 "0e age can be used in this context, Figure 4.15 pre­ a.. sents a graphical report from PARADE (a 0.90 Primavera Systems Inc. product) regarding the 0.85 progress of a project made up of a number of Concrete 3.031 . work packages. The performance curves on the 0.80 graph show that the actual expenditure and 0 10 20 30 40 50 60 70 80 90 100 progress on the project falls behind the planned Percent complete curve (baseline). The project is obviously be­ hind schedule and over cost. The tabular summary of costs at the There are two types of PFs: period PF bottom of the figure is presented in cost/sched­ and cumulative PF. The period PF is a short­ ule control system criteria (C/SCSC) format. term measure used for immediate control pur­ The budgeted cost of work scheduled (BCWS) poses, and action can be taken to remedy a is the cost baseline, the cost based on the origi­ disturbing trend. The cumulative PF is a long­ nal plan and the operating budget. The budget­ term measure of productivity used for forecast­ ed cost of work performed (BCWP) is the real ing the cost at completion. value of the work performed and is recorded The least common methods for evalu­ from reports compiled from the project site re­ ating productivity are input utilization tech­ flecting the actual progress and budgeted cost niques, i.e., activity surveys or time expenditure on each task. The actual cost of measurements, such as work sampling, foreman work performed (ACWP) is what has been ac­ delay surveys, and craftsman questionnaires. tually expended, regardless of the value of the The disadvantage of these techniques is that work. To analyze the performance of a project, more data on worker performance, in addition a set of variances is defined: the schedule vari­ to that provided by the costing system, must be ance (SV) given by SV = BCWP- BCWS and generated and that adds costs to the project. the cost variance (CV) given by CV = BCWP-

PRODUCTIVITY IN CONSTRUCTION With the aid of programs like P A­ Figure 4.15 Sample project performance report from RADE, project managers can input the required PARADE information regarding the progress of a project BCWS BCWP ACWP every month and acquire performance reports that can assist them in controlling the costs and

2000 schedule of a project. Whereas Figure 4.15 is a 100 1800 90 sample of such a report in graphical format, 1600 eo Figure 4.16 is a tabular report with the various 1400 70 indexes. The report encompasses the period, 1200 .. --- 60 1000 , cumulative, and forecasted-to-completion PF. " - 50 % 800 , " - 40 According to Figure 4.16, $1,140,000 600 " " 30 400 " "' 20 worth of work was scheduled for this period. 200 10 Only $182,000 was performed with an actual

JAN-- NOV DEC cost of $1,059,500. This yields an SV = 1993 -$958,000 and a CV -$877,500. Obviously an CUM BCWS 49 212 356 319 240 220 130 119 138 72 10 = CUM BCWP 104 61 12 0 0 unfavourable situation. CUMACWP 209 319 266 253 12 0 0 0 ETC 0 0 0 0 320 151 119 159 72 10 CUM BCWS 49 261 617 936 1176 1396 1526 1645 1783 1855 1865 1865 Reports can be by task, work pack­ CUM BCWP 104 165 177 182 182 182 182 182 182 182 182 CUMACWP 209 529 795 1048 1060 1060 1060 1060 1060 1060 1060 ages, or for the entire project. Through them, a LRE 209 529 795 1048 1380 1530 1650 i808 1880 1890 1890 project manager can pinpoint the problem areas Scale 1 : 1000 in a project. To illustrate, a report was obtained for Package 1.1 on this project (made up of four major packages each down to 10 sub-pack­ ACWP. (CV is the difference between the ages). The report given in Figure 4.17 shows monetary value of the work accomplished and that the performance curves for a particular the actual costs incurred.) work package are unfavourable and look si~i­ lar to those of the entire project. Summary of Abbreviations Others besides the cost engineer or ACWP = actual cost of work performed project manager should be concerned and BCWP = budgeted cost of work performed knowledgeable about productivity. When the BCWS = budgeted cost of work scheduled project engineer, superintendent, and foremen C/SCSC = cost/schedule control system criteria understand how productivity is measured and CPI = cost performance index how its results are interpreted, they can provide CV = cost variance more input into the control process. After all, EV = earned value they are the most familiar with the site and its PF = performance factor operations. The resulting information can be p-h = person-hour used when remedial actions must be taken to SPI = schedule performance index improve productivity. When productivity in SV = schedule variance cost/unit is not as high as management predict­ ed for a given work package, the superinten­ A schedule performance index (SPI) dent can provide the information required to is given by: assess the situation. An increase in cost/unit SPI = (BCWP/BCWS) x 100 value can be the result of factors quite apart from crews and their effectiveness. They could SPI reflects the efficiency of a task expressed be the techniques used, higher-than-usual as a percentage of EV. Low SPis require im­ hourly rates for craftsmen during the construc­ mediate attention or they will cause schedule tion season, or simply management practices, delays. Tasks with SPI over 100 are ahead of including poor planning. For the intermediate schedule. managers to be able to provide positive input into the corrective process and action plan, they A cost performance index (CPI) is must have an appreciation of the costing sys­ given by: tem. How the information flows, where it is initiated, how it is evaluated, and what can be CPI = (BCWP/ACWP) x 100 derived from it are essential factors in cost CPI represents the cost efficiency of a task ex­ communications and control. pressed as a percentage of the EV. Values below 100% indicate a cost overrun; over 100%, a cost underrun.

PRODUCTIVITY IN CONSTRUCTION Figure 4.16 Tabular report on project performance from PARADE

Page 1Aof 1B Page 1B of 1B

:Parade (R) Primavera Systems, Inc. Cost Performance Report- Work Breakdown Structure Report Date: I ;Title: Demo for Productivity in Construction Start Date: Period End Date: Report Period: 2 , r-·------~------~------~ Item 1 Current Period ' Cumulative to Date ' At Completion lr.. ------.... - - ... ------_._ ------... -- - ... - -- - - ... - --I------... - ...... ' , ; Budgeted Cost ' Actual ' Variance ; Budgeted Cost : Actual : Variance ' ; L 1 1 1 WBS Level • ' Cost !. ------.------• Cost 1 ------, atest 1 ~ , ~Work-- -,-Wo-rk-- - - ~ Work : , I Work , Work ; work :schedule, ' 'Revised • • 1 : i Scheduled ; Performed ; Performed ;Schedule; Cost i Scheduled ; Performed ; Performed ; ; Cost i Budgeted i Estimate : Vanance :

~ ..... ------.. - - ,------.------~ ------~ - --- - T ---I ------_.------,------,---- - +- -- -.· ... -- -- i - -- -- i------: 1Sample Parade Project 1 Jl : 1140.3 182.0 1059.5 -958.3 -877.5: 1189.0 182.0 1059.5 -1007.0 -877.5: 1865.0 1890.0 -25.0 : r------~------~------~ ------~ 1 ;cost of Money : • , j·------~ ------~------~ · ------~ - -~ ------~ ; Gen and Admin : .0 .0 .o .o .o: .o .o .o .o .o : .0 .o .0 ; : U-ndi~t~b-u~ed B~d~~t - - - ;x·x~x;;x~x~-x;x~;x~x~;x~;x;x;;x~;x;x·x;x~;x;x~;x~:~~;;x:;x;x·x;x:;x~x-x;x:;x;x:;x~;x~x;;x~;x;x:;x~;x~- - - - - .0 - - - - -.0 - -xxxx;xXx: r------·~------~------~------~------~ ;subtotal : 1140.3 182.0 1059.5 -958.3 -877.5: 1189.0 182.0 1059.5 ·1007.0 -877.5: 1865.0 1890.0 -25.0 : ~------· ~------j - ______1 ______~ ; Management Reserve :xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxkxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx~ .0 .0 .0 : ~ • _ - - _ - __ ..... __ • _ 1... • _ • • .. • - • • .... • ...... ______j ______... ______... __I ______...1 ;Total ' 1140.3 182.0 1059.5 -958.3 -877.5; 1189.0 182.0 1059.5 -1007.0 -877.5: 1865.0 1890.0 -25.0 : ~------~. (All Items in 1000) ------· ------.

Additional Readings Figure 4.17 Performance report by work package Adrian, J.J. 1979. Construction Accounting: BCWS BCWP ACWP ETC , Financial, Managerial, Auditing, and Tax. I Reston, Va.

100 Halpin D.W. 1985. Financial and Cost Control 90 Concepts for Construction Management. New , ...... -..... -' eo / 70 York: John Wiley and Sons. 5000 / / 60 / % 4000 / so / 3000 / 40 / / 30 2000 ,. , /.~· · ···-·~········ .. ·••······ .. ········· Earned 20 1000 ..,"' __ .. ~ 10 0 JAN-- FEB MAR APR MAY JUN JUL ~00 SEP OCT NOV OEC 1993 cuM sews 4B7 1799 3811 5514 6331 6B77 7263 7538 7650 7650 7650 7650 CUM BCWP 1041 1651 1767 1820 1B20 1820 1920 1820 1820 1820 1920 CUMACWP 1174 2605 3583 4504 4545 4545 4545 4545 4545 4545 4545 LRE 1174 2605 3583 4504 6874 7463 7738 7B50 7850 7850 7850

Scale 1 : 1000

PRODUCTIVITY IN CONSTRUCTION 5 Management Issues

5.1 Introduction Some companies in the construction industry, for example, the large Engineering Factors affecting productivity in con­ Procurement and Construction contractors, struction can be divided into two categories: aware of the importance of material manage­ human-related factors and management-related ment, have established computerized material­ factors. These factors affect the morale and tracking systems. They realized the need for motivation of individuals. better material management, especially for large complex projects that use thousands of Quality of supervision, material man­ components. The negative cost impact of ship­ agement, site planning, constructability, and ping delays and poor procurement procedures change management are the most significant became increasingly important to the project, management-related factors that influence pro­ and therefore these companies had to lead the ductivity directly. way in integrating good material-management 5.2 Quality of Supervision systems into their operations. Quality of supervision can be viewed The general building industry is begin­ ning to appreciate the importance of material in terms of leadership and team-building. These abilities create a positive environment management and the tremendous potential for increasing productivity and safety on construc­ for the individual worker. Everyone wants to tion projects. Smaller construction projects do be on the winning team. Good supervision has not require elaborate material management sys­ an obvious direct impact on productivity. tems. But regardless of size, some system, Workers can be demotivated with ill-informed, poor supervision, or ineffective supervision due whether manual or computerized, is necessary. to a high worker to supervisor ratio. Approxi­ As part of overall material manage­ mately 10% of the time on a project is spent ment, some database systems track the status of communicating instructions, and this can be major pieces of equipment and critical items. done effectively through good supervision. Spreadsheets are a convenient tool for tracking. Management and supervision are perceived by More comprehensive, integrated systems ad­ the worker as either competent, informed and dress all material-management functions for concerned, or not. This perception, if positive, both engineered equipment and bulk material. influences behaviour favourably and pays divi­ There are costs associated with the system cho­ dends through improved productivity. sen. Equipment, software, and training costs depend on the size and complexity of the sys­ 5.3 Material Management tem. Care must be exercised in selection of the "The construction industry lags far behind the system because of the costs and staffing re­ manufacturing industry in applying the con­ quired and to avoid costs disproportionate to cepts of materials management." (The Busi­ the size of the operation or company. For large ness Roundtable, 1983) projects with thousands of material items, ex­ tensive computerization is necessary; for small­ More recently the construction indus­ er projects, manual methods or spreadsheets try has become cognizant of the importance of will suffice. the management of project materials and equip­ ment, which can amount to 50% or more of Material handling is a significant com­ project costs. An estimated labour productivity ponent of material management and studies gain of up to 6% may be attainable through im­ have shown it to be a large percentage of site proved material management. Traditionally labour. In a series of 22 productivity studies labour productivity receives the most attention (O'Brien, 1989) carried out in Ontario, it was because the productivity of direct work can be found that mechanical and electrical tradesmen measured at a reasonable cost. Within the con­ were spending only 32% of their day on direct struction industry, there is a wide divergence in installation work, 20% on material handling, the degree of material management applied to 15% on indirect work, and the remaining 33% projects. No common methodology is used to on ineffective and miscellaneous operations. measure its effectiveness. Obviously many areas required improvement,

PRODUCTIVITY IN CONSTRUCTION ______.,

but the magnitude of material handling is espe­ 5.3.1 Material management steps cially noteworthy. A productivity improve­ ment program was instituted, which increased This section deals mainly with the at­ direct installation to 52%, and reduced material tributes of material management and the re­ handling to 12%. sponsibilities of those involved in performing them. A detailed understanding of each con­ Other studies show similar ratios of tributing function is required to comprehend direct to indirect work, with material handling the interfaces between material-management and waiting for materials amounting to a signif­ functions. A material-management system in­ icant percentage of the person-hours. These cludes the major functions of identifying, ac­ macro-studies highlight material handling and quiring, distributing, and disposing of materials management as opportunities for productivity at a construction site (CII, 1988). in1provement. The traditional approach to pro­ ductivity improvement has been to devote more By definition, material management is effort to the analysis of direct operations, such the management system for planning and control­ as cutting, assembling, and joining of compo­ ling all efforts necessary to ensure that the correct nents. A more effective method of improving quality and quantity of materials are specified in on-site productivity would be to reduce the per­ a timely manner, obtained at a reasonable cost, son-hours spent on indirect work, such as wait­ and available at the point of use when required ing for materials and material-handling. A (The Business Roundtable, 1983). worker should have the right materials at the Each firm has its peculiar material-man­ right time. For this to happen requires more agement system. Usually the responsibility for than good material-handling; it requires good the various activities has been spread between material management. engineering, purchasing, and construction. Some Material handling and movement can assign full responsibility and accountability to a be a hazardous activity. Most tradespersons are material manager; for most firms, the responsibil­ not trained in material handling, lifting, and ity is divided and therefore prone to problems. In fact, the more divided the responsibility, the transportation. Good material management will, through planning and control, improve more potential problems exist. productivity and also reduce risk by ensuring The steps in Figure 5.1 represent the that material handling is performed by trained process from identifying material needs to deliv­ and qualified tradesmen. Productivity must be ering the materials when required at the point of considered with the associated level of safety; use. They are only the key elements which make productivity and safety are closely related. up the whole material-management process. 5.3.2 Responsibilities Figure 5.1 Material management steps Responsibilities and authority of the Material Management Steps participants in the material-management process must be clearly established. An effi­ Sequence Contributing Action/Documents cient material-management system leads to im­ 1. Request for Quote (RFQ) Drawings, specifications proved productivity and necessarily includes all Material bills participants. The scope of each participant's Terms and conditions involvement must be clearly defined in contrac­ 2. Bids Approved bidders list tual documents. If not, increased effort will be Pre-qualification of bidders expended to rectify errors in quantity, quality, Bid evaluations or cost. Unexpected effort reduces productivity 3. Purchase Orders (P.O.) Bid clarification of the operation. Poor quality in the material­ Notice of award management process becomes apparent imme­ diately at the point of use. By comparison, 4. Expediting Vendor data Manufacturer inspection poor quality of engineering, for example, may Delivery not become apparent at all. Routings Figure 5.2 shows the contractual rela­ 5. Transport Carrier and route tionships and the key documents used to estab­ Ownership en route lish the scope of material management for each Customs participant. 6. Receiving Inspection and acceptance Receiving report If an owner purchases a long-lead item Storage and later assigns the purchase order to the con­ 7. Inventory Dispersal (material handling) tractor, a clear understanding of the purchase Inventory level order is required, as well as full knowledge of Surplus disposal any relevant correspondence, to ensure that nothing slips through the cracks.

PRODUCTIVITY IN CONSTRUCTION ______.,.

Although all major departments are in­ Figure 5.2 Relationships and key volved in a material-management system, the documents key departments are Engineering, Procure­ Request for ment, and Construction. Project needs are usu­ Quotes ally identified by Engineering, which also determines the specifications and quantities. Engineering generates a request for a quote (RFQ), which is completed by Procurement. Procurement develops a bidders list, solicits quotations or bids, evaluates bids with the assistance of Engineering for technical as­ pects, and issues a purchase order. Other de­ partments within Procurement expedite, inspect, and look after transportation. The Construction Department receives 5.3.3 Interfaces and their implica· the materials, inspects them upon receipt, tions for productivity stores, and ultimately issues them to the work stations. A material-management system con­ sists of numerous logical components, as Computerized systems are being in­ shown in Figure 5.3. As with any system, most creasingly used, especially for large projects. problems arise at the interface of functions, and The most important needs of these systems are management attention is therefore required. to provide a communication tool and to save The system includes the documentation and time on the execution of such functions as take­ procedures, as well as trained personnel to exe­ off and material lists. Any tool that satisfies cute the functions. these needs improves productivity by minimiz­ ing the cost to acquire materials and improving efficiency at the site, especially by the timely Figure 5.3 Construction material management delivery of the required materials. The level of Construction Materials Management sophistication depends on several circum­ stances, among them company size, and project Material Construction size and complexity. An organized program Engineering Control Procurement Supplier Construction Material System Control has a positive effect on staff and clients. Material-management problems can IEngineering I occur in the form of quantity and quality errors. Description Shortages disrupt the work pattern and require re-planning around the shortages. If the system I Material I Take-Off cannot detect shortages far enough in advance of material needs, the result is last-minute shuf­ IRequisition 1 fling of work crews. If the quality of materials and Bill of I _I Investigate_] Materials - ~ Supplier supplied is sub-standard, either the material is i-----···-. • Status • rejected or it is used, but requires additional ' ----,----·' J I person-hours to install. Rejected material must '+ --·I Purchase f Fabrication I Order 1 I be removed and replacement material handled; I I both operations require unanticipated person­ I • I hours. Products of inferior quality, such as a 14 -- - -1 Expedite f I - - lower grade of wood, may require additional I I ~ person-hours to accommodate excessive I r ·-----·-·: ,.I ___ warpage. (Beware of the dollar saved in pur­ · [Inspection f ~ Status - ----r----•' chasing; it may not necessarily be a dollar I' ' ' I' saved on the project.) I Shipping tH Receiving I ' r--:. ,...__~ Billing The most noticeable effect of poor l---- -1 f I Storage ~ material management is delays in delivery. t . I . These delays have an effect similar to errors in 'I 'Issue' for r I Installation ~ quantities, because the work flow is disrupted I ' and must be replanned. Disruptions cause lost I • I I j-4- - ---1Disposal l Surplus time and necessitate non-productive work to I I f ~. remedy the situation . r----1- --·-· . : Cost/Schedu le : - ---1' : Status • ------Those who have experienced projects ·------... .} where everything happens as planned will recall

PRODUCTIVITY IN CONSTRUCTION that the right material was delivered at the right 5.3.5 Material control time, and everyone associated with the project was aware that the project was well planned. The material-control function includes Morale rises on well managed projects. The determination of quantities, material acquisition, converse is true of poorly managed materials and distribution. The objective is to purchase programs, and the message to the workers is materials in a timely manner to avoid costly that management (or engineering) does not care. labour delays resulting from delivery delays and The result: productivity suffers. non-availability of materials. With an integrated material-manage­ Bills of materials are merged with mate­ ment system, materials are more likely to be rials specifications to establish quantities and available when needed and supervisors can quality for ordering. A milestone schedule plan the work around material availability. Re­ should be established for major items, so that a turning to a work area to replace a missing item complete plan is available (Figure 5.4). This wastes person-hours; this can be regarded as re­ plan includes the required-at-site date and final work. It has been reported that foremen spend issue of drawings and data for vendors, vendor as much as 20% of their time hunting for mate­ data, manufacturers' schedule, and delivery time. rials and another 10% of their time tracking purchase orders and expediting them (Bell and Field material control is required to Stackhardt, 1987). A material-management plan storage and issue of materials. A material­ system has a payback commensurate with the management system should provide an alert for amount and quality of the effort input. potential shortages. Control of inventory is re­ quired to prevent theft or unauthorized issue, 5.3.4 Preplanning and provides warranty protection for environ­ mentally susceptible items. Front-end planning is probably the sin­ gle most important determinant of a successful Various techniques can be adopted. A material-management effort (CII, 1988). Mater­ just-in-time technique requires careful planning ial management has to be an integrated activity, and a good system. By this method, materials with clearly established responsibilities assigned are bought for delivery just before they are to the owner, engineer, or contractor. Important needed, i.e., expenditures are incurred only decisions, such as site access and laydown when required and no sooner. This technique areas, schedule compression, cash flow restric­ yields significant cash flow benefits. Just-in­ tions, expenditure approvals and audit require­ time is better suited to large purchases but ap­ ments, are made early. All these decisions have plies to such bulk materials, for example, as an impact on costs and productivity. ready-mix concrete and asphalt paving.

Figure 5.4 Sample procurement milestone schedule

Legend l Sample Procurement Milestone Schedule Data sheeJs co_m[l.!ete 7 Pretil]l ve.nd,or datil] 2 RFO Issued a Certmef! ~@____, - f Receive blils 9 RAS date.______, Company XYZ -~ Bid tab compieie __10_ ETAjobsile Re-vamp project Contractor Co. 5 Eng~ rele~se PR -!· Mechanical Procurement Schedule Status to 90 08 20 L_EO awar~d ___ __j

June I July August September October November December FQmcast Vendor Req'd Bidders 1 Week beginning Monday 1 8 15 22 291 6 13 20 27 3 10 17 24 31 7 14 21 28 5 12 19 26 2 9 16 23 30 7 14 21 28 De l. Promise AI Site (R=Received) I ETA RAS (N=No bid) (S=Successful Week Number 1 2 3 4 5 6 7 8 9 14 1516 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1~ 1 ---;-;-;-; Blddat) - P-3306AIB M.W .H. LJGuld Ptnnps P New Impeller ! ~ phur _Gua!ti Bed __ AlP ------14WkS - X·B32 SGB Steam Pre·Healor P 3 45 6 8 10 Shop (Dim and Cor Checks) Suphur Guard Bed NP 3 45 6 _] ~ 10 v.6a7 & V·688 Reactor and Cl~y Filter P 3 45 10 Jet A Marox (Supply Mat'l) NP 1 12 wks I 1t 3 45 10 V·659166Q16561667/672 Misc. Vassel W01k P 2 45 6 10 6 wks JetAMerox (SupptyMaJ'IJ NP+---~---~ 2__ 3 5 6 10 Catalyst & Acetic Acid Educ1or ~ Hold Jel A Merox /VP B.S.-1/213/4/5/6(7/8 Basket Strainers P 3 45 6 10 8 wks Jet A Merox /VP Catalyst & Acel1c Ac1d Add111on Drums P - Hold JeiAMerox NP P-262 Condensate Pump --p· 2 3 45 6 8 12 wks New JeiAMerox NP 11 2 3 45 6 8 10 P·256 CausUc Pump --P- 2 3 45 6 0 !10wks J JetAMerox NP j 12 3 45 6 8 ~tO

PRODUCTIVITY IN CONSTRUCTION ______..

Another approach is the inventory handling is the key to improving productivity buffer approach, which can be costly from the and safety. standpoint of cash flow and losses due to theft. In a more extreme case, it can result in in­ Material handling can be categorized creased storage and double-handling which, in by the following five sections: tum, may increase costs and lower productivi­ • containerization and packaging ty. It is, however, unrealistic to expect every • movement to site item to arrive at exactly the right time. Some • off-loading at site and storing buffer is required, depending on the material • horizontal movement and complexity of the job. Most foremen want • hoisting and vertical handling. to be assured that all materials are available for Containerization and packaging. a particular operation, to avoid reassignment of Containerization and packaging require careful crews because of shortages. planning and organization. Various types of Trade-offs are necessary between just­ pallets, containers, and protection are available. in-time and inventory buffers. Often the inven­ The sequence of packaging or loading is impor­ tory buffers are a form of insurance to provide tant, especially for congested high-rise con­ continuous and unimpeded operations. The struction sites. more material stored at a site, however, the The use of skid-mounted equipment more double-handling is required. and equipment modules reduces on-site labour. 5.3.6 Procurement Modules require considerable preplanning and, because of the up-front engineering and manu­ Procurement includes purchasing of facturing effort, numerous person-hours are materials, equipment, supplies, labour, and ser­ transferred from the field to a shop environ­ vices for a project. Associated with the pur­ ment. Overall the result is improved labour chasing are the related activities of tracking and productivity on the site and lower project costs. expediting, routing and shipping, inspection and acceptance, handling, and storage and dis­ Movement to site. Movement to site posal of surplus. Procurement can be grouped usually involves trucking but can include rail, under three categories: the procurement of ma­ ship or air transport. Planning of the arrival of terials, labour, and subcontracts. shipments at the site is important so that crews and equipment can be available when required. Four cost categories (Barrie and Unplanned shipments result in waiting time for Paulson, 1992) must be considered to optimize the trucker or a deployment of workers to han­ the procurement of materials for minimum cost, dle the shipment. The most efficient methods and to some extent these same considerations of material movement could require winter apply to the procurement of labour and subcon­ roads, wide-load permits, or partial load restric­ tracts. These four cost categories are purchas­ tions. For special size loads, route planning is ing, shipping, holding and shortage costs, and necessary to avoid stalled shipments. trade-offs between the categories must be opti­ mized to achieve minimum costs. In these dis­ Off-loading at site and storing. cussions, the maximization of productivity is Unloading at sites requires trained material­ equated to the minimization of total costs. handling crews with the proper handling equip­ ment. A void handling material several times. On large projects it is common prac­ tice to produce a subcontract schedule and a Horizontal movement. Horizontal procurement schedule for major pieces of movement methods depend on the material equipment. For example, Figure 5.4 is a por­ being handled. Trucks and trailers are the usual tion of a schedule taken from an actual refinery conveyances but conveyors and cranes are also modification project. A scrutiny of the equip­ common. Insufficient handling equipment, ca­ ment plan reveals milestone dates for the com­ pacity or size have obvious negative effects that pletion of key steps, such as issue of the RFQ, lower productivity. purchase orders, and vendor drawings, bid re­ Hoisting and vertical handling. Ver­ of contracts, and required-at-site quests, award tical movement and hoisting require material delivery. and personnel hoists, cranes or other lifting de­ 5.3. 7 Material handling vices. Several decisions are required in plan­ ning the equipment for a job. The required A large percentage of site labour ac­ capacity, most suitable type, i.e., mobile, tivity involves material handling. As stated crawler or fixed hoisting equipment, and best previously, in the Ontario studies (O'Brien, location on the site, are decisions that have a 1989), 20% of the labour was initially for mate­ direct impact on project productivity. rial handling until a concerted effort was made to reduce this to about 12%. Reducing material

PRODUCTIVITY IN CONSTRUCTION ------~

For placing of concrete, which method is best? The choice could be a concrete pump Figure 5.5 Cost influence of deci­ or a tower crane. The better choice may be the sions as a function of project tower crane, considering that it will be required phase anyway to handle formwork.

Economic Conceptual Working Construction CommiSSioning Studies Design Drawing s, On a large bridge project, hoists cost­ Tender & ing $200,000 were installed to reduce travel High Award time on site. The payback period was estimat­ ed to be 9 months for a project of 18-month du­ ration. Numerous examples can be cited where productivity was improved through the use of proper equipment. In planning the construction Cost of the 72-story First Bank Tower in Toronto, Impact the owner-developer studied the good and bad points at various construction sites. Workers were each losing 3 to 4 hours a day at some sites because of the time required to get to and from their work stations. The owner-developer implemented a well conceived factory-like sys­ Low tem for moving men and materials. The Conceive Develop Define Execute Finish around-the-clock use of elevators was planned and priorities were established. Productivity Project Phases for handling materials was increased 600% for the installation of marble, 400% for electrical wiring, 260% for glass, and 800% for drywall. Constructability enhances the effec­ The innovation saved 1.33 million p-hs. tiveness of construction. It is a macro-produc­ tivity factor that should be the way of thinking Space requirements are usually at a of the entire project organization. It is manage­ premium because several trades may require ment action (at all levels) that creates this cul­ the same space. At different times, many key ture; not a separate function, but an on-going decisions relating to space, egress, and access process. It can be a motivator to the worker are made; all these decisions affect productivi­ when the 'smart' details or methods are used. ty. Considerations are traffic movement at the site, proximity of buildings and obstructions, For example, Figure 5.6 (a) shows de­ types of roads, turning space, and parking- just tails for a beam bearing on a masonry wall. to name a few. Tight dimensional tolerances are required for the beam holes to match the anchor bolt loca­ The concepts of material management tion. This alternative would require an accura­ are essentially the same for large or small pro­ cy of construction that is costly and jects; the differences are a matter of degree in unwarranted. Figure 5.6 (b) shows a detail that the areas of organization and staffing, documen­ better accommodates construction tolerances tation, vendor relations and computerization. with a lower installed cost. Material management has improved consider­ ably in the last decade, with the development of better tools and methods resulting in improved Figure 5.6 Beam bearing detail cost effectiveness through better productivity. 5.4 Constructability Constructability means making opti­ mum use of construction knowledge and expe­ rience in planning, engineering, procurement, and field operations to achieve overall project objectives (CII, 1986). It is the effective and timely integration of resources and technology into the early phases of the project and then maintaining the involvement.

Maximum benefits accrue if all stake­ holders- including the owner and contractor­ with construction knowledge and experience participate early in the project, and remain in­ volved. Figure 5.5 shows how decisions in the (a) Anchor bolts grouted early stages have the greatest cost impact.

PRODUCTIVITY IN CONSTRUCTION ------c. To enable efficient construction, de­ signs must avoid complex details and shapes so that they permit flexibility in construction methods and material substitutions. The design schedule must support the construction field­ work sequence. Good quality drawings, speci­ fications, and site information improve productivity. Drawings are frequently criticized for lack of clarity and content, forcing field crews to devise their own solutions. This transfers part of the design function to the site, which is costly, disruptive, and inefficient. Those un­ dertaking the dimensioning should consider construction needs and not scatter the dimen­ (b) Bearing plate grouted with anchors on U/S sions over several drawings. plate. Beam field welded to plate With vendors and suppliers, con­ structability is enhanced by timely engineering data, pre-assembly, shop-testing of compo­ 5.4.1 A traditional problem nents, and the provision of lifting lugs. As the construction process has Constructability is enhanced by stan­ evolved and become more sophisticated, the dardization, such as the use of manufacturers' separation of the design and construction func­ standard dimensions, standard steel connection tions has increased under the traditional form of details, piping assemblies, off-the-shelf electri­ construction procurement. Traditionally the cal and mechanical equipment. Designs can _be owner hires an engineer/architect who designs standardized to realize the benefits of duplica­ the facility. Construction is awarded to a con­ tion, symmetry, and repeatability. For exam­ tractor who procures material, labour, and ple, if the form work for every member is a new equipment and executes the contract require­ experience, costs will skyrocket. Montreal's ments. This method, which results in the sepa­ Olympic Stadium and the Sydney Opera House ration of functions, is primarily responsible for in Australia are classic examples of large pro­ any lack of constructability. A return to the jects where the cost of form work went out of master builder concept is heralded as a step to­ control. Numerous examples exist where the ward more efficient and cost-effective projects. use of modular design can reduce costs. Con­ crete formwork and house construction are ex­ 5.4.2 Constructability concepts amples where wastage could be minimized through modular design. During the conceptual phase of project planning, project objectives must be estab­ Structural constructability considera­ lished, major construction methods selected, tions include the use of such elements as pre­ sites chosen, and a contracting strategy devel­ cast staircases in high-rise cores. Straight oped. reinforcement bars, prefabrication of cages, and Overall project schedules must be con­ detailing of reinforcement to suit pour heights, struction-sensitive. A sequence of activities are cost-effective steps. must be established with realistic durations to Effective design and construction re­ prevent costly overtime, schedule acceleration, quire that construction expertise be utilized early or counter-productive high levels of labour for in the project schedule. Constructability im­ craftsmen. provement is possible with construction-sensi­ Major construction methods should be tive designs and construction-driven schedules. considered during basic design. Special meth­ 5.5 Change Management ods include prefabrication, pre-assembly, and modularization. Projects are characterized by change. A change typically results from a revision to Effective site layouts can facilitate project scope or to the details of construction, construction activities and reduce costs. Ade­ and the rework required to rectify errors. quate storage spaces, access and roads, with particular emphasis on clearances for operating Changes have a ripple effect on the equipment and traffic flows, must be provided. project, causing disruptions and delays. Con­ The use of permanent facilities and utilities sider the sequence of events that occur due to a should be investigated. single change. The project manager is notified

PRODUCTIVITY IN CONSTRUCTION ., ------of the change. This information is communi­ A gradual deterioration of the plan­ cated to the foreman, who then has to stop su­ ning and scheduling will occur with an increas­ pervising or planning the other work. The ing number of changes. The introduction of workers are informed of the change and are changes is similar to the addition of a new ac­ moved to another work activity. Later, when tivity to the scope of work, and this often re­ full details of the change and revised materials quires schedule acceleration. are available, the work will be restarted. Dur­ ing the entire change process, additional effort Some of the impact of changes is read­ is required from supervisory, management, and ily apparent; some is not. The management of other support functions. change has a major impact on productivity. The simple change, for example, may Additional Readings be to relocate a door; this is readily quantifi­ able. There are, however, unseen impacts of ASCE 1976. Civil Engineering, American the change that are disruptive and time-con­ Society of Civil Engineers, August 1976. surning. Time is lost during the scramble to re­ Barrie, D.S. and B.C. Paulson. 1992. Profes­ locate and re-instruct the workers for a suitable sional Construction Management. 3rd ed. New substitute task. York: McGraw Hill. Because of the interdependency of Bell, L.C. and G. Stukhart. June 1987. "Cost construction operations, changes affect the pro­ and Benefits of Materials Management." ASCE, ductivity of other activities that are not a direct 113 (2). part of the change. This can also influence labour productivity in the form of the learning Construction Industry Institute (CII). 1985. At­ and unlearning impact discussed in Chapter 3. tributes of Material Management. The Univer­ sity of Texas at Austin, SD-1. Figure 5.7 (Revay Report, 1991) illus­ trates the loss of productivity because of Construction Industry Institute (CII). 1986. change orders on mechanical and electrical "Constructability: A Primer," The University work. Similar figures exist for civil and archi­ of Texas at Austin, Publication 3-1, July 1986. tectural work. Construction Industry Institute (CII). 1988. Of the three curves shown, the lower Project Materials Management Primer. The curve results from changes only. Additional University of Texas at Austin, Publication 7-2. major causes of productivity losses have, as shown, a cumulative negative effect. Drucker, P.F. 1974. Management: Tasks, Re­ sponsibilities, Practices. New York: Harper Disruptions and delays affect produc­ and Row. tivity because of the stop-and-go of the opera­ tion, work being out of sequence, repetition of Leonard, C.A., 0. Moselhi, and P. Fazio. 1991. the learning cycle, unbalanced crews and fluc­ "Impact of Change Orders on Construction tuation in staffing levels. The work force can Productivity." Can. 1. Civil Engineering 18 (3). become demotivated because the managers and O'Brien, K.E. 1989. Improvement of On-Site supervisors are perceived as incompetent and Productivity. K.E. O'Brien and Associates, indifferent. Toronto, Canada. Revay and Associates. January 1991. The Figure 5. 7 Loss of productivity due to changes - Revay Report 10 (1), Toronto, Canada. electrical and mechanical work The Business Roundtable. 1983. "Modem 60 Management Systems." Report A-6. BRT 50 200 Park Ave., New York, NY, 10166. ~0 ;:. ·:;: 40 ~ :::l "0 30 ec. 0 rn 20 rn 0 _j 10

0 0 10 20 30 40 50 60 Change Orders, %

PRODUCTIVITY IN CONSTRUCTION ------Cit

6 Conclusion

6.1 Macro· Versus Micro· ally, as quality and productivity improve, both Productivity the contractor and the end user are winners. It is a win-win situation. It is important to differentiate between macro- and micro-level productivity factors in The production function produces order to analyze cause/effect relationships and data, which are analyzed and provide feedback take appropriate action. Macro-level factors for action toward improvement. The cycle con­ that influence the effectiveness of construction tinues until the required level of productivity is are those that often attract considerable rhetoric reached. An analysis of productivity is com­ but not enough specific actions or economic plete only when quality and safety are also support. Many talk about the need to improve considered. The construction industry must be­ productivity because it is the key to economic come serious if it is to improve in quality, safe­ survival. There is abundant scope for industry ty, and productivity. Supervisors and trades­ and government actions to enhance and pro­ men must continue to improve their efficiency mote an atmosphere for sustained progress. which, of course, is the thrust of this book. Japanese industries have developed a model for cooperation and effective support. The individ­ Figure 6.1 Productivity improve­ ual Japanese worker is not more productive ment as a continuous process than a North American worker; it is their sys­ tem that is more productive. Government, in­ dustry, and the financial community must eventually cooperate to provide synergistic sup­ port at a macro-level. In Canada very little money is spent on research or improving pro­

ductivity in construction. The industries that Modification reinvest sufficient resources to remain competi­ Input tive will survive and therefore the industry and construction companies must continue, and in­ deed increase, their contributions at a macro­ level. If the construction industry, which constitutes approximately 15% of the GDP, does not improve, foreign competition will con­ tinue to make inroads into Canada's traditional 6.2 Miscellaneous Ideas for markets. Improving Productivity These are the macro-level concerns, in Construction which must be addressed, but are beyond the scope of this book. However, the individual Construction is a unique industry construction company or person has an obliga­ where field experience plays an important role tion to improve productivity at a micro-level. in maintaining high levels of productivity. The The efficiency of labour and methods are the qualities of the field-management team will foundation for competitiveness and more effort eventually determine the levels of productivity must be paid to the measurement of productivi­ achieved on a given project. An experienced ty. It may not be possible to separate the indi­ person can tell about the level of activity on a vidual effects of all influences. But this should project in a variety of ways. For example, a not deter efforts to quantify their effects and unique and inexpensive technique for measur­ impacts on construction efficiency. Productivi­ ing productivity is observing the sound level on ty improvement and measurement of the effects a project. Not that a noisy project is a produc­ of related factors must become part of the daily tive project; however, a silent project is a non­ construction routine. productive one. The trained and experienced ear can detect when a job is moving well by the Productivity improvement is a contin­ productive 'hum' or tempo to a project: the roar uous process, as shown in Figure 6.1, and an of the crane every 5 or 6 minutes, the sound of integral part of total quality management. Ide- welders arcing, or the fleeting of an air tugger.

PRODUCTIVITY IN CONSTRUCTION ------~

Topics that were not covered in the pensive air travel. Telecommunications facili­ previous chapters, yet can affect productivity, tate solving design-related problems by provid­ include job security, safety, and advanced ing computer linkage between site personnel telecommunication. All successful construc­ and the engineer or architect. Similar systems tion firms - large, medium, and small - practise are in use for the review of drawings and other the job security theory. Job security can signif­ contract documents, providing instantaneous de­ icantly affect the tradesman's efficiency for ob­ cisions, thus improving productivity and reduc­ vious reasons. ing cost. Other applications include CAD-based crane-planning systems which improve con­ Safety issues were briefly discussed in struction productivity through faster and im­ Chapter 3. One consideration, perhaps under­ proved engineering and planning. Multimedia emphasized, is the effect of a serious accident use in training is catching on. This facilitates on productivity. A project that is moving along explaining to the site crews how a certain mater­ well and encounters a serious accident or fatali­ ial or equipment can be installed. Walk-through ty, never gets back to normal or regains its programs have also reached the personal com­ original tempo. Installing the proper safety puter market. A complex 3-D design of a plant program and taking safety seriously can help can be tested for constructability using such avoid such a scenario. programs prior to constmction, thus reducing New technological innovations can be chances of redoing work and effectively en­ hancing overall productivity. put to work in construction. This can greatly enhance productivity as it provides timely infor­ When measuring productivity, it is im­ mation, reduces travel to remote sites, and facil­ portant to know where the project stands. Im­ itates immediate corrective action in case of proving productivity combines the scientific emergencies. Examples include the closed-cir­ understanding of the issues affecting productiv­ cuit television communication systems that are ity, supplemented by work experience. being used to conduct meetings between engi­ neers in remote project locations and the main The construction industry should office. In addition to timely information and ac­ make better use of automation to improve the cess to the main office's immediate expertise, planning and control processes to remain com­ this eliminates routine, time-consuming, and ex- petitive, especially in a global market.

PRODUCTIVITY IN CONSTRUCTION