YUSQ -QU ALIT Y ANDS T AND ARDIZA TION ASSOCIA TION OFSERBIA &MONTENEGR O

UDK 658.5 YU ISSN 1452-0680 No 3 Vol. 33 2005 Management fr P I n a om 30thMa t p e MEN YUSQ -QU e r MAN n r s T a

t ADžMENT T i s o ALIT e O n EX c a o Y ANDS '' Specialedition l n

y t - T d A

W o 1s dv T r AL QU AND e CELLENCE o INTERN A anced v O r ARDIZA t June,2005,atBelgr i k T e ALNIM KV i w n GEMENT & g e TION ASSOCIA A

d and TION

C a o n AL JOURN Int n d ALITET f

e elligent TION OFSERBIA&MONTENEGR p r e r ALIT e n s c AL OM &IZVRSNOS e e n

t Appr e ade, Ser ' d ' T

o a oaches'', t t a

l Y

t h bia. e Q O

u T T a held h l i i t r y d

ASSOCIATION SERBIA & MONTENEGRO FOR QUALITY AND STANDARDS Kneza Miloša 9, 11000 Beograd, Serbia, Phone/fax: ++ 381 11 323 55 15

INTERNATIONAL JOURNAL

TOTAL QUALITY MANAGEMENT & EXCELLENCE MENADžMENT TOTALNIM KVALITETOM & IZVRSNOST

'' Special edition ''

Papers second reviewed and presented at the Third International Working Conference ''Total Quality Management – Advanced and Intelligent Approaches'', held from 30th May to 1st June, 2005, at Belgrade, Serbia.

UDK 658.5 YU ISSN 1452-0680 No 3. Vol. 33 2005 INTERNATIONAL JOURNAL „TOTAL QUALITY MANAGEMENT & EXCELLENCE” Vol. XXXIII, No.3, 2005

FOUNDER AND PUBLISHER: Association Serbia and Montenegro for Quality and Standards (YUSQ), Belgrade

EDITOR IN CHIEF: Prof. Dr Vidosav D. Majstorović, mech. eng. Mechanical Engineering Faculty, University of Belgrade, Serbia

INTERNATIONAL EDITORIAL BOARD: Prof. dr Guenter ARNDT Prof. dr Gunnar SOHLENIUS University of Wollongong, Wollongong, Royal Institute of Technology, Stockholm, Australia Sweden Prof. dr Daniel BRISAUND Prof. dr Dragutin STANIVUKOVIĆ University of Grenoble, Grenoble, France FTN, Novi Sad, S&M Prof. dr Barrie DALE Prof. dr Herbert OSANNA UMIST, Manchester, England Technical University, Wien, Austria Michel DEBENHAM Prof. dr Tilo PFIEFER Institute of QA, London, England RWTH Aachen, Aachen, Germany Sofija DJURDJEVIC Prof. dr Hendrik VanBRUSSEL YUQS, Belgrade, S&M Katolike University Leuven, Heverlez, Belgium Prof. dr Noriaki KANO Prof. dr Shu YAMADA Science University of Tokyo, Tokyo, Japan Science University of Tokyo, Tokyo, Japan Prof. Dr. Laslo MONOSTORI Prof. dr Albert WECKENMANN Hungarian Academy, Budapest, Hungary University Erlangen, Erlangen, Germany Prof. dr Vidomir PAREŽANIN Prof. dr Ton ven der WIELE ASTRA grupa, Beograd Erasmus University Roterdam, The Mr Zoran PENDIĆ Netherlands Institut Vinča, Beograd

TECHNICAL EDITOR: Siniša M. MARKOVIĆ, Dipl. mech. eng.

PROOFREADING Printing from authors' disketes

EDITOR’S ADDRESS: Association Serbia & Montenegro for Quality and Standards, Serbia and Montenegro, 11000 Beograd, Kneza Miloša st. 9, Phone/fax: ++ 381 63 33 10 15 E-mail: [email protected] Web page: www.jusk.org.yu

This edition financially aided by the Ministry for Science and Environmental protection of the Republic of Serbia. Manuscripts and illustrations not returned. The Journal is exempted from taxes, Decree No.413.304/7502 of 4. June 1985, passed by the Republican Secretariat for Culture. The texts published in this edition of the JOURNAL cannot be copied or printed without prior agreement of the author or the publisher.

Layout – A S&M QS, Belgrade Copies printed: 1000 Belgrade, September 2005.

ii

PREFACE

The Third International Conference "Total Quality Management - Advanced and Intelligent Approaches" was held from 30th May to 1st June, 2005 at Belgrade. During three days the Conference attended 241 participants from 18th countries.

Selected papers presented at this Conference are now offered to you in the special edition of the TQM&E International Journal. The International Special Edition Editorial Board reviewed the papers and after corresponding additions and corrections they are included in this edition. The papers reflect the world level in theory and practice of different TQM aspects today and as such may be used for education of experts for quality.

The main messages stated during plenary sessions and at the round table are: (i) the support for bi-annual holding of this Conference in Serbia, (ii) presented papers, particularly those from abroad, are a good basis for work on implementation in our country, (iii) the papers from Serbia & Montenegro (the majority) should be in future more oriented towards the results of the implementation, and (iv) the support for national plan of activities for quality improvement in 2005/6.

The Four International Conference with the same global topic and new topics will be held from 27th to 30th May 2007 at Belgrade.

Welcome in Belgrade 2007.

Prof. Dr. V. Majstorović, Conference Chairman

iii

CONTENTS

01 - INTELLIGENT FLEXIBLE DISASSEMBLY AND RECYCLING OF USED PRODUCTS TO SUPPORT TOTAL QUALITY MANAGEMENT AND SUSTAINABILITY IN EUROPEAN INDUSTRY …………………...... … 01 ¾ Keynote paper P. H. Osanna, M. N. Durakbasa, H. S. Tahirova, Department for Interchangeable Manufacturing and Industrial Metrology, Vienna University of Technology, Wien, Austria

02 - COMPUTER-AIDED DESIGN OF EXPERIMENTS FRAMEWORK - A COMPREHENSIVE APPROACH TO PROCESS IMPROVEMENT……………………………………………….…………………. 05 ¾ Keynote paper Albert Weckenmann, Peter-Frederik Brenner, Chair Quality Management and Manufacturing Metrology, Erlangen University Erlangen-Nuremberg

03 - A MANAGEMENT CONTROL PERSPECTIVE FOR QUALITY MANAGEMENT: AN EXAMPLE IN THE AUTOMOTIVE SECTOR………………………………………………………….....……… 09 ¾ Keynote paper Jos van Iwaarden and Ton van der Wiele, Erasmus University Rotterdam, The Netherlands

04 - THE VALUE OF ORGANISATIONAL CULTURE AND THE ROLE OF COMPETENCIES IN DELIVERING QUALITY PRODUCTS AND SERVICES……………………………………………….…….. 15 ¾ Keynote paper Michael Debenham, Professional Affairs Manager, Institute of Quality Assurance (IQA), London, United Kingdom.

05 - IMS IN THE M(E)SS WITH CSCS……………………………………………………………….……..……. 19 ¾ Keynote paper Professor Stanislav KarapetrovicAuditing and Integration of Management Systems Research Laboratory, Department of Mechanical Engineering, University of Alberta, Edmonton, Canada

06 - A CONTRIBUTION TO THE DIGITAL QUALITY CONCEPT RESEARCH……………….…… 27 ¾ Keynote paper Professor Vidosav D. Majstorovic, Dr.Sci., Mech. Eng., Mechanical Engineering Faculty, Belgrade, Serbia

07 - ORGANISATIONAL SUSTAINABILITY MANAGEMENT THROUGH MINIMISED BUSINESS EXCELLENCE MODELS……………………………………………………………………...... ……… 33 ¾ Keynote paper Rickard Garvare1) , Raine Isaksson2), 1) Luleĺ University of Technology, Luleĺ, Sweden, 2) Gotland University, Visby, Sweden.

08 - QUALITY MANAGEMENT MATURITY AND MANAGEMENT ATTITUDE TO QUALITY ...... 41 1)Mr Milan Ivanović, 2) Prof. Dr. Vidosav D. Majstorović, 1)LR, Belgrade 2)Mechanical Engineering Faculty, Belgrade

09 - DESIGN OF A SYSTEM FOR INTELLIGENT DATA-POINT PRE-PROCESSING IN REVERSE ENGINEERING……………………………………………………………………………………….....…… 45 ¾ Keynote paper I. Budak a), M. Sokovic b), J. Hodolic a), a) Faculty of Technical Sciences, University of Novi Sad, b) Faculty of Mechanical Engineering, University of Ljubljana,Slovenia

iv 10 - „FAST FORWARD” – DESIGNED PROCESS OPTIMISATION……………………………..…. 51 Prof. Dr.-Ing. R. Schmitt; Dipl.-Ing. J. Dören, Fraunhofer Institute for Production Technology, Department Metrology and Quality Management; Aachen, Germany

11 - INTERNET BASED AUTOMATION OF THE PROCESS "DESIGNING - MACHINING" OF A WORKPIECE……………………………………………………………………………………………………...... 57 Prof. Dr. K.-D. Bouzakis(1), Assistant Prof. Dr. A. Vakali(2), Lecturer Dr. G. Andreadis(1), E. Karapidakis(2), 1 Laboratory for Machine Tools and Manufacturing Engineering, Mechanical; Eng. Dept., Aristoteles University of Thessaloniki, Greece; 2 Computer Science Dept., Aristoteles University of Thessaloniki, Greece

12 - DEVELOPMENT OF AN INTELLIGENT MODULE FOR DECREASE OF MEASURING ERROR ON CMM………………………………………………………………………………..…………………………. 61 M. Stevic1, J. Hodolic1 S. Vukmirovic2, 1Insitute for Production Engineering, Faculty of Engineering, University of Novi Sad, Novi Sad, Serbia and Montenegro; 2Department of Control Systems and Automation, Faculty of Engineering, University of Novi Sad, Novi Sad, Serbia and Montenegro

13 - TRACEABILITY OF MEASUREMENT RESULTS IN INDUSTRY…………………………..…… 67 Prof. Dr. Bojan. Ačko, University of Maribor, Faculty of Mechanical Engineering, Maribor, Slovenia

14 - SYNERGY IN APPLICATION OF PRACTICES OF THE INTEGRATED PRODUCTION SYSTEM TOWARDS ACHIEVING MANUFACTURING EXCELLENCE……………………….…… 73 ¾ Keynote paper L Sukarma, Agency for the Assesment and Application of Technology, Indonesia

15 - CONTINUOUS IMPROVEMENT – A PREREQUISITE FOR SUCCESS…………..…..……. 81 ¾ Keynote paper R. Pendić1, V. Majstorović2, Z. Pendić3, 1 Eurosystems Group – Matheos Invest Group, Belgrade, S&M; 2 Mechanical Engineering Faculty, Belgrade University, S&M; 3 Institute of Nuclear Sciences “VINCA”, Belgrade, S&M

16 - ACCELERATING PERFORMANCE IMPROVEMENT THROUGH APPROPRIATE RESOURCE ALLOCATION……………………………………………………………………………….…………….. 87 L Sukarma, Agancy for the Assessment and Aplication of Technology, Indonesia

17 - APPLICATION OF TAGUCHI MODEL FOR QUALITY PRODUCT IMPROVEMENT....… 93 Mech. Eng. B.Sc. Tatjana V. Sibalija1, Prof. Dr. Vidosav D. Majstorovich2, 1 STMicroelectronics Malta Ltd., Industry Road, Kirkop, Malta, 2 Mechanical Engineering Faculty, Laboratory for Production Metrology and TQM, Serbia

18 - USING PARTIAL LEAST SQUARES REGRESSION TO ANALYSE QUALITY IN HIGHER EDUCATION…………………………………………………………………………………………………………..……. 101 ¾ Keynote paper L. Catellani, Bianca M. Colosimo, Q. Semeraro, Dipartimento di Meccanica – Politecnico di Milano, Milano (Italy)

19 - RESEARCH AND DEVELOPMENT OF DIGITAL QUALITY MODEL IN SCM……….…… 109 ¾ Keynote paper 1) Nenad Stefanovic, Zastava Automobiles, 2) Vidosav Majstorovic, 3) Dusan Stefanovic, 1)Information Systems Division, Kragujevac, SCG 2)Mechanical Engineering Faculty, Belgrade, SCG; 3)Faculty of Science, Kragujevac, SCG

v 20 - ISO MANAGEMENT SYSTEMS STANDARDS AND SOCIAL RESPONSIBILITY CONNECTION: (NOT QUITE) JOINED-UP OPINIONS OF ISO’S STAKEHOLDERS…...…… 119 Pavel Castka 1), Michaela A. Balzarova 2), 1) University of Canterbury, Christchurch, New Zealand, 2) Brno University of Technology, Brno, the Czech Republic

21 - RESEARCH INTO THE POSSIBILITIES OF THE APPLICATION OF THE FUZZY LOGIC TO THE DEVELOPMENT OF THE QUALITY EVALUATION MODEL………..…… 125 Marko Mirkovich, M.Sc.1), Prof. Dr. Dragan Radojevich2) , Prof. Dr. Vidosav D. Majstorovich3), Prof. Dr. Janko Hodolic4), 1 Bauxite mines,s.c. Quality & Information Centre ,Niksic 2 Mihajlo Pupin Institute , Belgrade 3 Mechanical Engineering Faculty , Laboratory for Production Metrology and TQM ,Belgrade 4 FTS, Production Mechanical Engineering Institute, Novi Sad

Appendix

1. FINAL REPORT FOR CIRP: Third International Working Conference “TOTAL QUALITY MANAGEMENT – ADVANCED AND INTELLIGENT APPROACHES”……………………………….. 129

2. Fourth International Working Conference - FIRST ANNOUNCEMENT - CALL FOR PAPERS, PRESENTATIONS AND PARTICIPATION …………...... ………...... …………….. 151

vi Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

INTELLIGENT FLEXIBLE DISASSEMBLY AND RECYCLING OF USED PRODUCTS TO SUPPORT TOTAL QUALITY MANAGEMENT AND SUSTAINABILITY IN EUROPEAN INDUSTRY Keynote paper P. H. Osanna, M. N. Durakbasa, H. S. Tahirova Department for Interchangeable Manufacturing and Industrial Metrology Vienna University of Technology, Wien, Austria

Summary: Important charge of the environment, reduced availability of natural resources and the increasing growth of waste require new concepts and strategies to recycle technical consumer goods as there are household instruments, consumer electronics and passenger cars - instead of land filling, burning or steel production a high potential of recycling is necessary. In view of the large quantity and high personnel costs, an advanced disassembly technique is needed working more rational than traditional manual processes and cheaper than highly sophisticated fully automated highest technology machines and devices. One of the biggest problems of the used and to be recycled or disposed products besides the growing mass of waste is the complex composition. An economic, sustainable and ecological procedure is necessary to prepare and to refurbish products, therefore the disassembly is the first step of a high potential of recycling. Keywords: Disassembly, waste, life cycle assessment, sustainability, recycling, consumer products.

1 INTRODUCTION European “Take-Back Law” requires automobile manufacturers to take back all vehicles which were ever In modern industrial production the very important sold in that country. Voluntary agreements have been issues of the protection of environment and sustainability widely accepted by industry and the threat of legislation must be taken into consideration increasingly /1, 2/. has subsided slightly. Manufacturers of products for every day life - consumer Essential advantages to application of old automobiles electronics, automobiles, house hold devices - are facing are established by the following directives of EU and increasing requirements of consumers, public opinion and according to their transformation to the national right: governments to minimize the pressure on environment - • The directive 75/442/ European Economic increasing consumption of raw materials and energy, Community of advice from July 15, 1975, about growing pollution and waste. waste. The complexity of scrap recycling from the above • The directive 96/61/ European Economic Community mentioned consumer products is a strongly of advice from September 24, 1996, about the interdisciplinary field where there are items like integrated evasion and reduction of environmental collection logistics, disassembly, components re-use, contamination (IPPC - the instruction). recovery of precious and rare materials like copper, • The directive 1999/31/ European Economic recycling of non-hazardous materials and disposal of Community of advice from April 26, 1999, about hazardous and toxic substances. All these require the gathering waste. coordinated work of interdisciplinary teams. European Union Environment Council approved in • The directive in 2000/76/ European Economic 1990 the Commission’s Strategy for Waste Management, Community of the European parliament and advice which included ELVs (End of Life Vehicle) as a Priority from December 4, 2000, about combustion of waste. Waste Stream. • The directive in 2000 / 53 / European Economic Concerning of the Waste Strategy Priorities the Community of the European parliament and advice generally accepted hierarchy of priority with respect to from September 18, 2000, about old automobiles. waste is: The overall targets for the improved process, which • prevention, are in line with those proposed by the EU, are: • recovery, • for a car being scrapped in 2002, a maximum of • disposal. 15% of initial weight to go to landfill • for a car being scrapped in 2015, a maximum of 5% Within the term recovery the concept includes the of initial weight to go to landfill reuse of parts, the recycling of material and the recovery of energy. Economic viability is a necessity for the The Objectives of the European Commission Proposal implementation of the improved disposal process under for Automobiles are the following: market conditions. If analysis of detailed monitoring - Avoidance of waste, information indicates that an action is viable, but direct - Reduction of landfill demand, and financial support is required for it to be sustainable, then - Reduction of toxicity. the means of providing such support will be considered.

1 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Currently, the European Commission is proposing a that is a maximum of 5 % disposal. directive in which the vehicle manufacturer has the prime responsibility for the product. However, the complex nature of vehicles, the length of life, and the other industrial sectors involved subsequently, point to a shared 2 LIFE CYCLE COSTS responsibility between all affected parties. Essential targets in accordance with item 7 of the EU In the past the cost structure of a product contained instructions 2000/53/EG about old vehicle are /3/: only cost of development, production including quality • Target from 2006: assurance, marketing, sales and service. As this is mainly 85 % recycling influenced by the manufacture of the product it is 80 % material recycling important to take into consideration the Life Cycle Costs that is a maximum of 15 % disposal. what means that the real costs in each life stage are added • Target from 2015: (see Figure 1). 95 % recycling 85 % material recycling

Figure 1: Cost Influences During Product Life Cycle

For this purpose a detailed quantitative assessment of 3 END OF LIFE MANAGEMENT expenses for a product has to be carried out taking into account the following: The challenges for innovative product design are two - Development costs: fold: on the one side having to contain material and Roughly 5 % of the life cycle costs originate from this energy flows within the product life cycle to close the stage. On the other side to this small part of the costs loop, on the other side reducing the overall consumption between 60 % up to more than 80 % of the total life cycle of materials and energy to make the loop thinner (see costs are fixed during the development stage. The Figure 2). These both should be achieved whilst at the influences of other stages on the cost level are small same time the consumer needs are to be identified and compared with those of research and development. met. The closed life cycle consists therefore of two main - Production costs: streams: In any cases one can operate on the rough rule which - The "traditional" product life chain (such as production, allows the determination of production costs based on distribution and use). Introducing environmental concerns costs of different materials used for engineering purposes in this stream is called "Eco-design". - Use costs: - Re-using and recycling products, components and The costs of the product use phase can be calculated materials (such as take back, re-use and recycling). on the basis of the energy costs per time unit that the Operation of this stream is called "End-of-life product is in use. Management". - Disposal costs: By disassembly of further on usable parts considerable The costs of take-back, recycling and final disposal reduction of use costs and disposal costs can be achieved. are comprised by the costs of take-back systems, In this respect it is very important that different branches recycling and final disposal processes as well as the of industry collaborate insofar that different experiences returns for the gained secondary raw materials. are exchanged. In order to account for the different points of time when costs occur, the costs for usage and recycling have to be discounted over the life span of the product.

2 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Figure 2: Closed Life Cycle of Consumer Products

4 Development of an Intelligent Disassembly - Disassembly Tools especially developed for robots System - Components Data Base including data of re-usable and re-manufacturable parts A fixed automated single purpose disassembly system - Storage Device for tools and parts cannot operate economically at the time being. So it is - Transport System and feeding system for products very important to design such a system with high as to be disassembled possible flexibility. Such a system can be called - Clamping Device and fixture system for parts with "Intelligent Flexible Disassembly Cell" (see Figure 3) and different geometry and dimensions it consists of different main modules: - Manual Disassembly Station for specific tasks - Disassembly Robot or handling device with special - Sensors for force, torque, visual recognition, features like path and force control position and distance - Robot Gripper for a wide spectrum of parts with - Intelligent Cell Control Unit able to process different geometry and dimensions information from extended sensors.

Figure 3: Modules of an Intelligent Flexible Disassembly Cell.

3 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Flexible semi-automated disassembly can only be carried out economically when the conditions are satisfactory. There must be enough products available at the disassembly cell and the separation technique must be appropriate.

5 SUMMARIZING AND FINAL REMARKS

It can be summarized that the modern society begins to realize that it is necessary to find economical and ecological solutions in order to reduce industrial waste and to preserve natural resources. It has been proved that the only solution was to recycle technical consumer products as there are household instruments, consumer electronics and passenger cars. In this connection, disassembly of products can be referred to as a first step towards efficient recycling and supports strongly sustainability and TQM in European industry.

RERFERENCES:

/1/ EN/ISO 14001: Environmental Management Systems - Specification with Guidance for Use (ISO 14001: 1996). /2/ ISO 14001: 2004: Environmental Management Systems - Requirements with Guidance for Use. ISO Standard, 2004. /3/ Directive 2000/53/EC of the European Parliament and of the Council of September 18, 2000, on end-of-life vehicles /4/ Daichendt, K., Kopacek, P., Zebedin, H.: A New Strategy for a Flexible Semi-automated Disassembling Cell of Printed Circuit Boards. Proceedings of International Conference ISIE'2001, Pusan, Korea, 2001. /5/ Perlewitz, H., Seliger, G., Tomiyana, T., Umeda, Y.: Activities of Japanese Industry for Product Recycling. Proceedings of 2nd International Seminar on Re-use, Eindhoven, NL, 1999. /6/ Penev, K.D.: Design for Disassembly Systems - A Systematic Approach. PhD Thesis, Eindhoven, NL, 1996. /7/ Bröte, S.: Disassembly Systems - Process Analysis and Strategic Considerations. PhD Thesis, Linköping, DK, 1998. /8/ Kopacek, B.: The European WEEE Concept - A Contribution to the Recycling and Re-use of Waste from End- of-life Electrical and Electronic Equipment in Europe. PhD Thesis, Vienna-Wien, A, 1999.

4 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

COMPUTER-AIDED DESIGN OF EXPERIMENTS FRAMEWORK - A COMPREHENSIVE APPROACH TO PROCESS IMPROVEMENT

Keynote paper Albert Weckenmann, Peter-Frederik Brenner Chair Quality Management and Manufacturing Metrology, Erlangen University Erlangen-Nuremberg

Summary: The paper proposes the structure of a computer-aided Design of Experiments framework, which supports the user in the application of Design of Experiments (DoE) for continuous process optimization. The computer-aided DoE framework includes an advisory module based on a hypertext system and a knowledge-based system for choosing the suitable experimental designs and several tools - called assistants - for supporting the user in system analysis and planning of experiments. The objective of the framework is to strengthen the employees skills in DoE and to use this powerful quality management technique extensively for continuous process improvement. Key words: Computer aided, Process improvement, Framework.

INTRODUCTION extraneous variables Total Quality Management (TQM) is a management approach of an organization centered on quality based on the participation of all its members and aiming of long term success through customer satisfaction, and benefits inputs output to all members of the organization an to society /5]. One s key element of TQM is to use quality management tools and techniques for continuous process improvement /2]. Black Box However, most quality management tools and techniques process model only utilize a small part of their full potential. More theoretical advanced quality management tools and techniques must be transformed into easily accessible methods to achieve popularity. A powerful, but complex quality management Figure 1: Process model technique for continuous improvement of manufacturing processes is Design of Experiments (DoE) /1], /3], /6]. In literature, Design of Experiments is described as The application of DoE requires trained and skilled follows: employees. To improve the application of complex • /ISO 3534-3] quality management techniques, it is necessary to develop “The arrangement in which an experimental computer-aided tools to support the user. By means of program is to be conducted, and the computer-aided DoE tools, employees can strengthen selection of the levels (versions) of one or their methodical skills in DoE and the user is assisted in more factors or factor combinations to be designing experiments. included in the experiment." Definition of Design of Experiments • For an accurate definition, refer to /4] Design of Experiments is mainly used during process “A designed experiment is a test in which planning, manufacturing and assembly. some purposeful changes are made to the In these phases of the product life cycle, a model of input variables of a process or system so the existing processes is usually used to explain why that we may observe and identify the Design of Experiments should be applied. Assignable reasons for change in the output response." causes (inputs) and extraneous variables (stochastic inputs) affect the response variables (outputs). The cause- The procedure followed in DoE differs from a series and-effect relationships within the process are partially or of experiments as they are conducted in practice. completely unknown (Fig. 1). Normally, one influencing parameter is changed and the In Design of Experiments, inputs are understood to be effect on the response variables is observed. The change assignable causes that affect the response variable. The to the influencing variables is therefore somewhat response variables are outputs that describe the observed intuitive. If the required result is not achieved by result of experiments. An unintentional and changing one parameter or if a further improvement is uncontrollable influencing variable is termed an extra- expected with a certain setting, another influencing neous variable. The influencing variables used for the variable is varied. The results are optimum response experiments are also called factors. variables obtained largely by chance. The path there is no longer reproducible. It is not known which influencing variable really caused the

5 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 improvement in the response variables. In design of • to conduct acceptance tests on machines, work pieces experiments, an attempt is made to change several and processes. influencing variables at once with a systematic procedure and to represent their effect on the response variables in a Methodical approach of Design of Experiments reproducible way. The methodical approach in Design of Experiments The goals of DoE are as follows: can be divided into five phases: system analysis, • to determine assignable causes that have a significant experimental strategy, realization of experiments, effect on response variables, evaluation and validation. For all of these phases, specific • to obtain information about the project, process, and DoE knowledge must be provided to gain new machine, information about the process or the product (Fig. 2). • to optimize the quality of the products and processes,

DoE knowledge Process information

- Ishikawa diagram - factors - Brainstorming System analysis - noise variables - interactions

- full factorial designs - fractional factorial designs Experimental strategy - central composite designs

- randomization Realisation of experiments -experimental data

- calculation of effects - significant factors - analysis of variance Evaluation and interactions - regression analysis - mathematical models - determination of the confidence interval

- optimal settings of - possibilities of extension Validation factors of the experiments

Figure 2: Phases of Design of Experiments

The aim of the phase system analysis consists of During the phase realization of experiments, the discovering process or product variables, called factors, planned experiments are conducted /4]. Within this phase, which influence the process yield output or the product the factors have to be adjusted with respect to the specifications. The identification of the factors is done determined levels and the experiments have to be run in a with the help of systematic quality management tools like careful way. Ishikawa-Diagrams, Brainstorming and Metaplan- The phase evaluation deals with the mathematical Technique. analysis of the experimental results. The focus lies on the The phase experimental strategy deals with the application of mathematical and statistical methods as selection of an appropriate experimental design. For this, well as the graphical analysis of the effects. The objective the DoE user must have a deeper knowledge of the is to investigate optimum settings of the factors as characteristics of the different experimental designs like process parameters for achieving the optimized output. full factorial or fractional factorial designs. Furthermore, In the phase validation, confirmation experiments relevant factors with a supposed impact on the process have to be run in order to confirm if the proposed factor output or the product quality have to be chosen and levels settings are right. This includes the documentation of the have to be defined. new information concerning the product or the process.

6 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Without the mentioned DoE knowledge it is not based on the already known process and product possible for a worker or an engineer, responsible for a information. process or a machine, to plan, run and evaluate • the user should be advised when choosing experiments in a meaningful way. For the user it is very adequate experimental designs which fulfill the objective important to know how to select the proper DoE methods of the planned experiments as well as in selecting of to conduct meaningful and economic experiments as well mathematical and statistical methods. Furthermore the as to apply the selected methods in very short time on the user should be advised in the application of adequate process. For this, it is necessary to develop user-friendly methods and tools available with quality management. tools which guide and assist the DoE user directly and • to guide the user through the different phases of “just in time” in all phases of DoE /7]. To supply the user DoE and to assist him in regard to his specific tasks when with the DoE knowledge “just in time”, the tools should planning experiments, so called assistants should be be computer-aided. included in the system. These assistants should help the user in planning experiments without having a big Requirements of a computer-aided Design of theoretical background in statistical tools. Experiments framework In general the compatibility of the system with existing software packages for DoE which are especially Regarding the above mentioned phases of DoE and designed for evaluating raw data of the experiments is the necessary knowledge, the following requirements of a required. That means that the data gained from the system computer-aided DoE framework can be pointed out: analysis phase and raw data from the experiments should • to apply the different techniques, methods and be transferred to statistical software packages via an tools of DoE in an efficient way, the user should have interface so that the analysis of the experimental results direct access to the DoE knowledge where it is needed. can easily be managed. Furthermore the computer-aided Furthermore, the user should be able to read up on the DoE framework must be capable to be integrated into an fundamental quality tools like the Ishikawa-Diagram in existing computer environment like CAx tools. order to use these tools for the system analysis. • process information obtained during the phase Concept of the computer-aided Design of system analysis regarding the process or the product like Experiments framework factors, interactions and noise variables should be documented consistently. Based on that primary process To provide the user with access to DoE knowledge information, cause and effect relationships between the i.e. directly on the shop floor, the computer-aided Design single steps of a process chain can be evaluated. In of Experiments framework is designed as a web-based addition the system should contain learning algorithms system. The user can access the system via a standard which help the user when drafting experimental strategies web-browser installed on a standard personal computer (Fig. 3).

user interface in web-browser

Beratung input output advisordes Benutzers

direct access integration of access to knowledge-based to DoE system process data-base for assistants knowledge information process linkingVerweis to auf Hypertextsystem information hypertext system

hypertext system

assistants integration in

hypertext documents for DoE knowledge

W E B S E R V E R

Figure 3: Concept of the computer-aided Design of Experiments framework

The main components of the computer-aided Design • Hypertext system: the features of a hypertext of Experiments framework are: system are nodes and links. Nodes can contain text, graphics, audio, video, animation and images while links

7 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 connect nodes related in a certain manner. The linking The data-base system allows the user to collect, to capability allows the non-linear organization of structure and to analyze process data and process information in the hypertext system. In the hypertext information which is needed for continuous system, the explicit knowledge concerning the different improvement. User guidance and user assistance during techniques, methods and tools for DoE is stored the phases of Design of Experiments (DoE) is very according to the five phases of DoE. The usage of a important for the meaningful application of experimental hypertext system allows the user to navigate through DoE methods for continuous process improvement. Because of knowledge on demand and just in time, via the web-based the minimizing of processing cycles in distributed interface of the system. process chains, the factor time becomes a critical success • Knowledge-based system: the knowledge-based factor for production systems and therefore for the system contains the formulated knowledge concerning application of DoE within complex process chains. the different types of experimental designs like full To optimize the yield output of process chains, the factorial, fractional factorial and central composite DoE user must apply the different techniques, methods designs. Via a web-based question-answer-dialogue the and tools in very short time. For this a computer-aided user is guided through a rule-based selection of adequate DoE framework was developed. The system includes an experimental designs step-by-step. For this, the user is advisory module based on a hypertext system and a asked to specify the number of factors and the number of knowledge-based system for choosing the suitable factor levels as well as pointing out the objectives of the experimental designs and several tools - called assistants planned experiments i.e. running screening experiments - for supporting the user in system analysis and planning or running optimizing experiments. The knowledge-based of experiments. In forthcoming work, the system will be system explains with the help of the hypertext system the extended with learning methods and the results will be recommended experimental designs. Furthermore the transferred to the industry. knowledge-based system is used to guide the user through the hypertext system with regard to his main interests. • Data-base system: in a relational data-base ACKNOWLEDGEMENTS system the user can store the results of the system analysis phase as the initial step of DoE as well as the The current work represents results of the research results of the experiments. The web-based access to this project “Design of Experiments for Manufacturing data base allows the user to get relevant process Processes of finely structured Electronic Products with information directly on the shop floor. Significant factors Numerous Variations” of Collaborative Research Centre influencing the yield output and optimal process SFB 356 "Production Systems for Electronics" funded by parameter settings obtained by the planned experiments the Deutsche Forschungsgemeinschaft, DFG (German are documented consistently. Research Foundation). • Assistants: the assistants are realized as java- applets which are integrated in the hypertext system and are accessible directly via the web-interface. The purpose of the assistants is to advise the user in planning and REFERENCES designing experiments without having a strong statistical background. An assistant is implemented in the system, /1] Anderson, M. J.; Whitcomb, P. J.: DoE simplified. which helps the user to calculate the required sample size New York: Productivity Press Inc., 2000 to conduct experiments with high statistical power. /2] Dalken, B. G.; Cooper, C. L.; Wilkinson, A.: The proposed framework was realized for a process Managing Quality and Human Resources. Oxford: Blackwell chain in electronics manufacturing. The benefits of the Publishers Inc., 1997 framework are: /3] Fasser, Y.; Brettner, D.: Process improvement in • employees such as engineers are trained directly electronics production. New York: Wiley, 2003 /4] Montgomery, D.C.: Design and Analysis of where they apply the DoE knowledge, they have just experiments. New York: Wiley, 1991 in time and on demand access to the hypertext /5] Pfeifer, T.: Quality Management: Strategies, Methods, modules, Techniques. München: Hanser Verlag, 2002 • the DoE framework guides the user through the /6] Scheffler, E.: Statistische Versuchsplanung und - different phases of DoE, auswertung: Eine Einführung für Praktiker. Stuttgart: Deutscher • data from the process chain is collected, structured Verlag für Grundstoffindustrie, 1997 and analyzed. /7] Weckenmann, A.; Rinnagl, M.: Intelligent methods in Design of Experiments (DoE). In: Sebaaly, M. F. (Ed.): The American University in Dubai: Proceedings of the International NAISO Congress on Information Science Innovations (The International NAISO Congress on Information Science CONCLUSIONS Innovations ISI'2001 (Congress IQMM 2001) Dubai, U.A.E 17- 21. March 2001). 2001, pp. 983-989. The computer-aided DoE framework gives the user access to the DoE knowledge. With the hypertext system it is possible to learn about the different tools, methods and techniques according to the methodical approach.

8 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

A MANAGEMENT CONTROL PERSPECTIVE FOR QUALITY MANAGEMENT: AN EXAMPLE IN THE AUTOMOTIVE SECTOR

Keynote paper Jos van Iwaarden and Ton van der Wiele, Erasmus University Rotterdam, The Netherlands

Summary:Two important trends in current business climate are an increasing product variety for customers and shortening product life cycles (Pine, 1993; Da Silveira et al., 2001; The Economist, 2001). An increasing product variety can be seen in the ever-increasing supply of and demand for alternative products and services in the market place. These days, customers can chose from many different types, colors, flavors and sizes of products. At the same time product life cycles are becoming shorter in many industries because of more products being influenced by fashion trends and increased (global) competition. Increasing product variety and shortening product life cycles may have major implications for many management control systems. This paper studies their effect on one of these control systems: quality management. In order to manage quality, organizations typically aim to do three things: to cement relationship with customers (and other stakeholders), to reduce variation in key processes as much as possible, and to improve processes and products in a continuous step-by-step manner. So, quality management control systems are typically based on measures of customer satisfaction, reduction of variation and step-by-step continuous improvement (Dean and Bowen, 1994; Wilkinson et al. 1998; Handfield and Melnyk, 1998; Dale et al., 2000; Dale, 2003). However, the relevance and effectiveness of all of these could be influenced by increasing product variety and shortening product life cycles. The increasing speed of change may subject the classic step-by-step Plan-Do-Check- Act (PDCA) based performance improvement loops to major strain. Since an updated product or process may already be in place before any improvements projected can be implemented. Moreover, many of the traditional tools and techniques aiming at reducing variation assume large batches of the same or similar products that are repeated over time. But batches are becoming smaller and the likelihood that a process will be repeated in exactly the same form is decreasing (Von Corswant and Fredriksson, 2002). So, the possibility of variation increasing is occurring at the same time as the basic assumptions required for traditional reduction of variation are under attack. Therefore, many of the currently used quality management systems of firms are based on assumptions that are challenged by the two trends and it is questionable whether these quality management systems are still useful in the traditional format. The empirical part of this paper is based on case study research at European automotive companies. The automotive industry is interesting for a number of reasons. Firstly, it has from the beginning been leading edge in quality management (e.g. Toyota) (Dale et al., 2000; Womack et al., 1990). Secondly, increasing product variety and shortening product life cycles are already visible in the automotive industry (Pine, 1993; The Economist, 2001; Von Corswant and Fredriksson, 2002; Womack et al., 1990; Alford et al., 2000; Agrawal et al., 2001). Car manufacturers keep introducing new models at a high pace and the option lists for cars are getting longer, although many features that used to be options in the past have now become standard equipment. Life cycles are under pressure because sales drop rapidly after a few years of production and even face-lifts cannot do much to counter that. Thirdly, many automotive firms are sharing platforms with other brands in the same firm or with competing firms in an effort to retain mass production as much as possible. This indicates that manufacturers try to reduce complexity by sticking to traditional mass production as much as possible, while on the other hand they try to offer customers the experience of a unique car. Fourthly, current quality management systems are clearly under strain in the automotive industry since many product recalls are not caused by internal problems at the car manufacturers but they arise from problems at their suppliers and even at their sub-suppliers. This paper argues that quality management systems need to adapt to cope with increasing product variety and shortening product life cycles. Explorative empirical research by means of case studies at three European automotive companies gives an indication of how quality management systems could develop. Keywords: Quality Management, Automotive, Customer

INTRODUCTION management attention. It is clearly more straightforward to manage a single mass production process than to The impact of the two trends of increasing product manage a number of production processes with large variety and shortening product life cycles on product varieties. The unpredictability is caused by the organizations lies in their ability to increase complexity constant flow of new product introductions and product and uncertainty. If companies have to produce a growing updates. Because they imply that success in the market number of different products and on top of that have to place may last only shortly (until your competitors constantly introduce new versions of all these products, introduce new versions of their products). Any business they are operating in a complex and unpredictable environment is a mixture of stability (predictability) and environment. The complexity is caused by the large instability (adaptation to changes) (Prater et al., 2001). number of different processes that all require Yet increasing product variety and shortening product life cycles are moving many firms towards more

9 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 unpredictability and instability. A survey among quality Beliefs Boundary experts led to the conclusion that quality management has Systems Systems to change radically in the short term, and that instantaneous response to changing market demands will be the single most important challenge of the future for Core Risks to be Values Avoided quality management (Mehra et al., 2001). Consequently, to study the effects of increasing product variety and shortening product life cycles on quality management, a model is needed that can Business distinguish between, on the one hand, simple and stable Strategy environments and, on the other hand, complex and unpredictable environments. Existing quality models like the quality award and business excellence models (e.g. Critical Strategic Malcolm Baldrige National Quality Award, European Performance Uncertainties Business Excellence Award, and the Deming Prize) are Variables not appropriate for this purpose because they do not make this distinction. Therefore, it is needed to look for an Interactive Diagnostic appropriate model outside the quality field. Control Control Systems S On the basis of a literature review it can be concluded ystems that quality management consists of three core building blocks, which are (1) customer orientation, (2) process Figure 1: Simons’ four levers of control model (Simons, 1995) control, and (3) continuous improvement (Dean and Bowen, 1994; Wilkinson et al. 1998; Handfield and Beliefs systems Melnyk, 1998; Dale et al., 2000; Dale, 2003). Based on Beliefs systems are used to inspire and direct the these three building blocks it can be argued that quality search for new opportunities. A beliefs system is the management can be seen as a control system because all explicit set of organizational definitions that senior three building blocks aims to control an organization’s managers communicate formally and reinforce processes and to improve and change these processes in systematically to provide basic values, purpose, and response to changes. Therefore, a logical place to look for direction for the organization. The definitions espouse the a model is in the field of management control. values and direction that senior managers want subordinates to adopt. These core values are linked to the business strategy of the firm. A formal beliefs system is created and communicated through such documents as MANAGEMENT CONTROL credos, mission statements, and statements of purpose. Boundary systems A model in the field of management control that can Boundary systems are used to set limits on distinguish between, on the one hand, simple and stable opportunity-seeking behaviour. Boundary systems environments and, on the other hand, complex and delineate the acceptable domain of activity for unpredictable environments, is Simons’ four levers of organizational participants. Unlike beliefs systems, control model (Simons, 1995). This model is shown in boundary systems do not specify positive ideals. Instead, figure 1. Simons’ four levers of control model is used to they establish limits, based on defined business risks, to balance control mechanisms in an organization in order to opportunity seeking. realize the business strategy. The model distinguishes Diagnostic control systems four different types of control mechanisms: (1) beliefs Diagnostic control systems are used to motivate, systems, (2) boundary systems, (3) diagnostic control monitor, and reward achievement of specified goals. systems, and (4) interactive controls systems. Two of Diagnostic control systems are the formal information these four levers increase individual freedom (i.e. beliefs systems that managers use to monitor organizational systems and interactive control systems), and two restrict outcomes and correct deviations from preset standards of individual freedom (i.e. boundary systems and diagnostic performance. These feedback systems, which are the control systems). backbone of traditional management control, are designed to ensure predictable goal achievement. Three features distinguish diagnostic control systems: (1) the ability to measure the outputs of a process, (2) the existence of predetermined standards against which actual results can be compared, and (3) the ability to correct deviations from standards. Interactive control systems Interactive control systems are used to stimulate organizational learning and the emergence of new ideas and strategies. Interactive control systems are formal communication systems managers use to involve themselves regularly and personally in the decision activities of subordinates. Based on the unique strategic

10 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 uncertainties they perceive, managers use these systems The same reasoning that applies to general control to activate search. Interactive control systems focus systems also applies to specific control systems, like attention and force dialogue throughout the organization. quality management. Therefore, quality management They provide frameworks, or agendas, for debate, and systems can also be classified according to the four levers motivate information gathering outside of routine of Simons’ model. Moreover, the two mentioned trends channels. These control systems stimulate search and of increasing product variety and shortening product life learning, allowing new strategies to emerge as cycles are expected to have a major impact on quality participants throughout the organization respond to management. So, a shift is expected from a major focus perceived opportunities and threats. An interactive on diagnostic quality management systems to a more control system is not a unique type of control system: important role for interactive quality management many types of control systems can be used interactively systems. by senior managers. The four different control levers in the model of Simons and their relation to strategy are summarized in table 1. METHODOLOGY

Table 1: Relating the four levers of control to strategy To empirically test the hypothesized shift towards an increased importance of beliefs systems and interactive control systems in a situation of uncertainty and Control Control of Purpose Communicates system strategy as complexity, three case studies have been conducted at European automotive manufacturers (OEMs). Because of Empower the explorative nature of this research, a case study Beliefs and expand Vision Perspective approach is a suitable methodology (Yin, 2003). systems search In both organizations interviews have been held with activity the quality manager, supply chain manager, logistics Provide manager, production manager, and human resources Boundary Strategic Competitive limits of systems domain position manager. On top of that three relevant first tier suppliers freedom have been selected for each of the OEMs. At each of Coordinate these suppliers interviews were held with the account and manager for the OEM (in some cases together with the Diagnosti monitor the quality manager of the supplier). c control implementa Plans and goals Plan All interviews were conducted by two interviewers systems tion of and each interview took between 1.5 and 3 hours. The intended interviews inside the OEM organization were all focused strategies at the changes in management systems that have taken Stimulate place over the last ten years in the field of responsibility Interactiv and guide Strategic Pattern of e control of the interviewee. The interviews at the suppliers emergent uncertainties actions systems focused on the changes that have taken place in the way strategies the relationship between the supplier and the OEM is Source: Simons, 1996, p. 304 managed by the OEM. Each interview was written down by both

interviewers and, based on these two write-ups, a final Any control system in an organization can be write-up of the interview was produced. Out of these classified according to the types that Simons write-ups the most important quality management issues distinguishes. The four different types of control systems and developments were derived by means of discussions work together to realize the business strategy. To be able about the write-ups by a group of academic quality to do this successfully, there should be a balance between experts. The derived issues and developments were the different types of control systems. If there is too much presented to the interviewees during a discussion meeting focus on just one or two types of control systems, the in which these managers could express their perceptions organization may have difficulties in realizing its and opinions. strategy. The right mix of control systems depends partly on environmental factors like the predictability and complexity of the market in which the organization is RESULTS operating. If the environment is predictable and not complex, an organization can put more emphasis on the diagnostic control systems and boundary systems. The results of the case studies at the three OEMs have However, if the environment is unpredictable and been interpreted by means of the Simons model by complex, a stronger focus on beliefs systems and placing the identified quality management issues and interactive control systems is necessary. In the current developments in the four levers. This means that each environment the two trends of increasing product variety issue or development has been judged by a group of and shortening product life cycles bring about academic quality experts and positioned in one or more unpredictability and complexity for many organizations. levers of the Simons model. Table 2 shows an example of

11 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 developments and issues in the area of process control in dashed arrows indicate a current development or a European automotive manufacturer. The numbering planned/expected future development for which already indicates the sequence of developments. Solid arrows some evidence has been found. indicate a development that has happened already, while

Table 2: An example of developments and issues in an OEM’s process control

Beliefs Boundary Interactive control systems Diagnostic control systems systems systems

1. Inspection of incoming

supplies

3. Establishment of direct communication between new product development teams at the OEM and development teams at the suppliers, to avoid quality problems during future production

2. Assessment and rating of the quality of the suppliers

and use of these ratings for procurement decisions

The developments in table 2 are as follows. Over the knowledge and experience of these suppliers to improve years the importance of the supply chain has increased for the design. this manufacturer. Parallel to that there has been an Because the sequence of developments in time is evolution in the management and control of the quality of known, it becomes clear which levers were important at the suppliers. More than five years ago this manufacturer what moment in time. So, the sequence of developments used incoming inspection to diagnostically measure the indicates which levers receive the most attention at a quality of the products that were delivered by its certain point in time. From table 2 it becomes clear that suppliers. So, in this stage the quality management the supplier focus of this manufacturer has shifted over system used was a diagnostic control system because it time from the right side of the matrix (i.e. diagnostic measured the compliance to preset quality standards. control systems) towards the left side (i.e. interactive When this manufacturer realized how time and resource control systems). consuming this policy was, it started to move towards However, in most cases these shifts do not mean that quality assurance by means of supplier assessments and a previous lever gets no attention at all once the focus is ratings. The quality performance of each supplier is on another lever. In most cases the manufacturers in this monitored on a day-to-day basis and recorded in a research kept existing quality management systems in supplier database. The ratings of suppliers are used in the place but felt the need to put more emphasis on different decision process that takes place when new supply kinds of quality management systems in order to achieve contracts will be given to suppliers. This is again a their quality strategy. quality management system that is diagnostic in nature Table 2 also shows that (for the example presented) because it uses predetermined performance measures that two levers of the Simons model (i.e. beliefs systems and lead to a ranking of suppliers from which the top boundary systems) receive no attention because there is performing suppliers will be selected for future contracts. no quality management system in place in the In recent years, a development has been started towards organization that controls the suppliers from the co-development between this manufacturer and its perspective of these two levers. The absence of certain suppliers. This entails that there is communication and levers in the management of the quality of the discussion between the manufacturer and its suppliers manufacturer’s suppliers is clearly demonstrated by the about new products. So, the manufacturer shows its plans matrix in table 2, which allows managers to think about for future products to its suppliers and it asks these the consequences of missing levers. In the example in suppliers to comment on the plans in order to tackle table 2 there is no boundary system. So, the manufacturer possible quality problems while the product is still in the has imposed no clear boundaries for the quality level of design phase. In an interactive way, the manufacturer its suppliers. Apparently top management thinks that such discusses its new designs with suppliers and uses the boundaries are not necessary because the ranking systems

12 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 that are in place ensure that only the best performing July 2001. suppliers are awarded new contracts. Therefore, it may /4/ J.W. Dean, and D.E. Bowen, “Management theory and not be necessary to define minimum quality levels. total quality: Improving research and practice through theory However, the boundary systems do not only define development,” The Academy of Management Review, vol. 19, no. 3, pp. 392-418, 1994. minimum levels of quality but also maximum levels. /5/ A. Wilkinson, T. Redman, E. Snape, and M. These maximum levels may be just as important as Marchington, Managing with Total Quality Management, minimum levels because superior quality may come at a Theory and Practice. London: Macmillan Press, 1998. high price. So, although it may be desirable to have the /6/ R.B. Handfield, and S.A. Melnyk, “The scientific highest possible quality, this may not be the smartest theory-building process: a primer using the case of TQM,” strategy from a cost perspective. In the automotive Journal of Operations Management, no. 19, pp. 321-339, 1998. industry products need to last only for a limited number /7/ B.G. Dale, M. Zairi, A. van der Wiele, and A.R.T. of years, so it is not necessary to manufacture car parts Williams, “Quality is dead in Europe – Long live excellence, that last for ages. It may be wise for automotive True or false?,” Measuring Business Excellence, vol. 4, no. 3, pp. 4-10, 2000. manufacturers to determine the relationship between the /8/ B.G. Dale, Managing Quality (Fourth edition). Oxford: technical lifespan of a car part and its costs. If the costs Blackwell publishers, 2003. go up if the technical lifespan is increased, the /9/ F. von Corswant, and P. Fredriksson, “Sourcing trends in manufacturer should question whether it is really the car industry, A survey of car manufacturers’ and suppliers’ necessary to go for the highest quality car part with the strategies and relations,” International Journal of Operations & longest lifespan. Production Management, vol. 22, no. 7, pp. 741-758, 2002. Based on the example in the previous paragraph, it /10/ J.P. Womack, D. Jones, and D. Roos, The Machine can be argued that the presence of quality management That Changed the World. Macmillan Publishing Company, issues and their developments over time in relation to the 1990. /11/ D. Alford, P. Sackett, and G. Nelder, “Mass different levers of the Simons model may give an customisation – an automotive perspective,” International indication of the quality maturity of the organization. So, Journal of Production Economics, no. 65, pp. 99-110, 2000. if a company has quality management systems in all four /12/ M. Agrawal, T.V. Kumaresh, and G.A. Mercer, “The levers of the Simons model, it may be more quality false promise of mass customization,” The McKinsey Quarterly, mature than a company that does not have quality no. 3, pp. 62-71, 2001. management systems in all four levers. /13/ E. Prater, M. Biehl, and M.A. Smith, “International supply chain agility, Tradeoffs between flexibility and uncertainty,” International Journal of Operations & Production Management, vol. 21, no. 5/6, pp. 823-839, 2001. /14/ S. Mehra, J.M. Hoffman, and D. Sirias, “TQM as a CONCLUSION management strategy for the next millennia”, International Journal of Operations & Production Management, vol. 21, no. In the current business environment the two trends of 5/6, pp. 855-876, 2001. increasing product variety and shortening product life /15/ R. Simons, Levers of control: How managers use cycles may have an impact on quality management of innovative control systems to drive strategic renewal. Boston: firms. This article argues that the automotive sector is Harvard Business School Press, 1995. /16/ R. Simons, Performance Measurement & Control experiencing the influence of the two trends. The research Systems for Implementing Strategy, Text & Cases. New Jersey: in this article shows how Simons’ four levers of control Prentice Hall, 1999. model can be used to categorize quality management /17/ R.K. Yin, Case Study Research, Design and Methods systems and indicate the direction in which changes are (Third Edition). London: Sage, 2003. needed This research has presented a novel application of Simons’ four levers of control model to the field of quality management. Based on the experience with three case studies at European automotive manufacturers, this approach seems to have potential. However, the methodology that has been used for this research needs to be developed further to improve its usefulness for scientific research and practical application.

REFERENCES

/1/ B.J. Pine II, Mass Customization - The New Frontier in Business Competition. Boston, MA: Harvard Business School Press, 1993. /2/ G. Da Silveira, D. Borenstein, and F.S. Fogliatto, “Mass customisation: Literature review and research directions,” International Journal of Production Economics, no. 72, pp. 1-13, 2001. /3/ The Economist, A long march, Mass Customisation, 12

13

Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

THE VALUE OF ORGANISATIONAL CULTURE AND THE ROLE OF COMPETENCIES IN DELIVERING QUALITY PRODUCTS AND SERVICES

Keynote paper Michael Debenham Professional Affairs Manager, Institute of Quality Assurance (IQA), London, United Kingdom.

Summary: This paper is not the result of academic research but has been developed as a result of observations by the author and feedback from various organisations that are affiliated with the IQA or organisations that simply have representatives who attend the various special interest groups supported by the author. It is an attempt by the author to remind the reader of the importance of people and the organisational culture that they build in delivering the organisation's vision and goals. Today the emphasis is on management system standards, models and awards and this is an attempt to redress the balance and highlight the importance of people who, after all, power systems and processes and without whom goals could not be achieved. Indeed it is people with the appropriate knowledge, skills and energy, working within a positive organisational culture, that really make the difference and can effect change within the organisation.

One of an organisation's key assets should be its personnel throughout the organisation. Thus it can be culture and the people that contribute to this culture. argued that the primary asset to the organisation is the There are some very sophisticated models to define an influence that can be exerted on human attitudes and thus organisational culture but this paper intentionally offers a behaviour patterns as individuals interact with customers, simple view of organisational culture. This simple view suppliers and colleagues alike. A secondary asset is the proposes that an organisation's culture is made up of the influence on trading partners and customers from a set of following elements that combine to support sound and fair organisational values and beliefs, either organisational purpose and deliver organisational goals: published or understood from the working relationships. • Organisational values and beliefs Clearly before embarking on a programme of change • Organisational structure it is important to establish the state of the current culture • Human attitudes and behaviour patterns within the organisation. It is not only important to These are visually set out below showing values and understand the current culture but also the culture that is beliefs coupled with the organisational structure or desired and the means to move from 'what is' to 'what system supporting human attitudes and behaviour needs to be'. patterns. How to determine 'what is?' There have been a number of techniques developed to measure organisational culture but perhaps the most effective is the 'nine factors' method developed by Jeff Cartwright and used in the UK (Cartwright 1999). Using the internal survey techniques outlined in this publication will give a good snapshot of the current organisational culture and begin to highlight the challenge ahead to move towards 'what needs to be'.

How to determine 'what needs to be'? The first criteria to be established are the set of organisational values to which the organisation will aspire. A description of what is meant by a learning culture, which is sometimes is described as a culture for opportunity, is outlined by Debenham in his paper 'Fit for the future -the evolving role of the quality manager'. It can be argued that such a culture is the basis for delivering sound product and service quality and the

effective improvement in an organisation's performance. Indeed, it is the organisational values cascaded A great deal of attention is currently being given to throughout the organisation via the structure that will models and standards in this regard but without a sound influence the attitudes and behaviour patterns of supporting organisational culture, management system models will not deliver in these key areas of business.

15 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

With regard to organisational values, management experience. Indeed how can the definition of competency will have to address the organisation's position in respect be applied to an organisation when competency to all or some of the following general issues. (sometimes also referred to as ability) is defined as the How does the organisation: application of knowledge, skill and experience? a) Do business with its customers? (Quality of Whether or not we accept that competencies can be product or service, restrictive practices, cartel, applied to an organisation, the value of defining bribery and related issues) 'organisational competencies' is that it firstly encapsulates b) Do business with its suppliers or subsidiary the vision (or standard) of our organisational culture, organisations? (Quality of product or service, secondly it provides the starting point for all subordinate exploitation of developing world labour and resources competencies and thirdly it demonstrates to individuals and related issues) how their individual competency links into the greater c) Compete with its competitors? scheme of things. d) Address its social responsibility? So by encapsulating the vision of our new e) Value its personnel? (Health and safety, organisational culture within the individual competencies employment, personnel development issues) an organisation can annually assess with its personnel f) Value the relationships between departments, how effectively the new culture is being accepted. It can groups and individuals? be further used to reward those who accept change and g) Value innovation, knowledge sharing and identify those who resist change and provide mentoring performance improvement? and training in these cases. h) Value sustainable processes and the environment? So how will organisations define organisational i) Maintain and replace its facilities, plant and competencies? This will of course be specific to each equipment? organisation but typically they will be defined under By developing the organisation's position in relation some of the following headings: to the above issues a model of the organisation's culture • Customer focus and relationship management can be envisaged and the reasons for implementing this • Leadership and management culture can be arrived at. Thus, the collective values and • Business development and innovation beliefs can begin to be established within the • Delivering outcomes management team. • Working with others and teamwork • People development How to achieve 'what needs to be?' • Performance improvement and learning We now need to address the question of how the • Organisational change values and beliefs that have been developed by management can be cascaded throughout the Let us consider an example of how the lower level organisational structure so that the attitudes and competencies might be developed from organisational behaviour patterns of individuals are changed to support competence of 'customer focus' using the IQA model. these values and beliefs. Customer Focus - the Vision a) Share and promote the vision We exist to serve our members and to place quality at b) Carefully define the drivers for the culture. Use the heart of organisations examples of how organisational culture enhances Customer Focus - the Organisational Competence business progress and performance improvement Achieving Customer Satisfaction: c) Example set by management The ability to meet the needs and expectations of our d) Select and motivate champions for change current customers at the same time as we are preparing to e) Identify resistance and educate deliver what our current and unknown customers will f) Develop organisational competencies that support want tomorrow. these values and beliefs and the supporting Customer Focus - Director Level Competencies individual competencies derived from the organisational a) The ability to establish and promote the climate and competencies culture of a customer-focused organisation g) Measure the change in organisational culture using b) The ability to communicate the vision and mission both competency assessment and internal throughout the organisation survey techniques c) The ability to evaluate performance in delivery of h) Reward adoption of values, educate and mentor product or service quality and initiate any where there is rejection of values necessary actions Customer Focus - Manager Level Competencies a) The ability to measure and improve customer satisfaction USING ORGANISATIONAL COMPETENCIES b) The ability to create and develop customer-focused TO DEFINE INDIVIDUAL COMPETENCIES teams c) The ability to develop a structure and culture to enable staff to take ownership of issues, deliver The concept of an organisation having a set of positive outcomes and learn from mistakes competencies may be considered a little unusual for those d) The ability to develop initiatives that keep the who have been used to thinking of competence in terms organisation customer focused of individuals applying their skills, knowledge and

16 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Customer Focus - Supervisor Competencies CONCLUSION The ability to deal with non-routine customer requirements and complaints finding the most Quality is everyone's responsibility and it is the effective solution culture of an organisation and the people with the The ability to monitor customer feedback and appropriate knowledge, skills and energy that deliver a satisfaction to ensure a consistently high standard of consistent quality of product or service to the customer. delivery Many organisations devote significant resources to The ability to anticipate customer needs within the development of management systems but relatively defined frameworks to ensure the customer is few commit equivalent or sufficient resources to satisfied developing the organisational culture or even establishing The ability to set and maintain a professional organisational values and this balance needs to be relationship with the customer which is mutually reviewed, as systems alone will not deliver a consistent beneficial and to guide others in this area quality of product or service. The ability to supervise and support others in meeting customer expectations Customer Focus - Support Staff Competencies The ability to provide accurate products, services or REFERENCES information requested by customers under contractual requirements or policy guidelines /1/ Cartwright, J, 1999 'Cultural Transformation. Nine The ability to use initiative to provide a professional Factors for Continuous Business level of service where customers are not Improvement'. Pearson Education Limited. straightforward /2/ Debenham, MGS, 2004 'Fit for the future - the The ability to use the most appropriate method to evolving role of the quality manager'. 4th communicate with customers throughout the International Quality Management Conference in Tehran. process The ability to taking ownership of tasks and inform others as appropriate The same approach can be used for the other organisational competencies that an organisation wishes to adopt. The value of using this approach is that firstly all individual competencies are aligned with the organisation's vision and secondly it enables individuals to see how their own competencies help to drive the organisation towards its vision.

MEASUREMENT OF COMPETENCY

Traditionally, competencies are classified as either hard or soft where hard competencies are those that can be measured objectively using a series of tests and soft competencies are those that are measured subjectively by carrying out an assessment at interview or over a period of time using observational and monitoring techniques. It should be added that hard competencies may well be measured using a combination of test and observational and monitoring techniques. Examples of hard competencies will be the ability to carry out a safety procedure, or a special work technique such as those needed by a lathe operator or a welder, data entry clerk or call centre operators. Examples of soft competencies are those listed above in the example under the heading 'Using Organisational Competencies to define Individual Competencies'. The purpose of measuring competency is so that any gaps between the specified competency and that possessed by individuals can be determined and measures (training, mentoring, etc) can be put in place to close the gaps that have been identified. It also enables an organisation to provide second parties with the assurance that critical activities are being undertaken by individuals with the appropriate competencies. 17

Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

IMS IN THE M(E)SS WITH CSCS Keynote paper Professor Stanislav Karapetrovic Auditing and Integration of Management Systems Research Laboratory Department of Mechanical Engineering, University of Alberta, Edmonton, Canada

Summary: Looking at the shear number of different national and international initiatives in the area, integration of management systems has evidently become one of the hotter topics of quality management research, but even more so, of the related practice. This paper discusses such integration from the perspective of international management systems standards, developed to address specific organizational functions or stakeholders. The environment and scope of integrated systems is analyzed first, followed by an explanation of the basic concept behind these systems, as well as their context and content. Subsequently, the assimilation of a new series of standardized complaint systems for customer satisfaction into the overall integrative framework is discussed. The paper concludes with a brief overview of expected future developments in integrated management systems. Keywords: Quality, Management System Standards, Integrated Management System, ISO 9000, ISO 10000

INTRODUCTION complaint systems (CSCS), namely ISO 10001-10003, follows. An example of possible integration of CSCS or 2 There are so many management system standards (CS) into an overriding MS is also discussed. Finally, (MSS) existing today, and possibly even more coming up further IMS initiatives are addressed. in the future, that the situation faced by the companies looking to implement them can be rightfully described as a “mess of MSS” or simply the “M(E)SS”. Apart from various industry-specific standards, such as ISO 16949 M(E)SS (automotive), ISO 22001 (food) and ISO 29001 (oil), the avalanche of industry-generic, yet stakeholder-specific Concept MSS includes ISO 9001 (quality), ISO 14001 (environment), OHSAS 18001 (health and safety), and Much like our cosmos, the MSS universe (i.e., the the recently-named ISO 26000 corporate social “M(E)SS”) was created in a “Big Bang” almost twenty responsibility standard. Each of these function-oriented years ago, and is constantly expanding in the four- MSS carries a whole series of additional supporting dimensional “Space-Time” continuum. Evidently, the standards. For instance, the quality management set MSS “Big Bang” occurred with the publication of the currently includes at least fourteen standards other than first ISO 9000 models, which contained a minimum set of ISO 9001, some of which, e.g., ISO 10002 for complaint elements required to establish a meaningful quality handling and ISO 10012 for measurement processes, can assurance system in any organization. Immediately after, be converted into management systems (MS) of their a natural expansion of both the content and the context of own. Making the M(E)SS hugely overwhelming is the these standards commenced, as ISO 9000 developers and fact that new MSS are being tirelessly cranked out by practitioners quickly realized the need for supporting both national and international standardization bodies. standards that provide more detail not only on the specific In order to somehow “resolve” the M(E)SS, two elements of such a generic system, but also on how to possible approaches are evident, namely the integration evaluate and implement the system itself. Consequently, of MSS, on one hand, and of MS, on the other. Realizing numerous guidelines on terminology (ISO 8402: 1987, the first idea, however, which amounts to a single MSS now ISO 9000: 2000), auditing (ISO 10011: 1990, now covering the full range of diverse organizational aspects, ISO 19011: 2002), and application (e.g., the former eight- would be largely useless (e.g., see Karapetrovic and part ISO 9004: 1994 series, now the single ISO 9004: Jonker (2003), for a detailed explanation). Instead, 2000), followed. standard users and developers alike are turning towards Contrasting the objectives (e.g. the stated purpose) the models and methodologies for the integration of against the outcomes (e.g. the resulting benefits) of the management systems (IMS). This paper therefore establishment of MSS models in companies provides provides an outline of the IMS solution to the M(E)SS. another insight into the reasons why the M(E)SS An illustration of the four dimensions that form the expansion is innate and the M(E)SS itself is needed. For M(E)SS, and consequently contain the components of an example, reporting the results of two large surveys of IMS, is given first, together with a discussion of some Spanish companies, Casadesus and Karapetrovic (2005A) relevant research topics. Subsequently, major alternatives illustrate eighteen different benefits of ISO 9000 in amalgamating MS models and requirements to create implementation, as perceived by the respondents. Two an IMS are addressed. A brief overview of a new series benefits clearly stood out from the field, namely of international standards related to customer satisfaction decreased product nonconformities and increased customer satisfaction, with at least 20% higher reports

19 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 compared to the next closest benefit in both surveys. e.g. ISO 14031: 2001 for performance measurement (e.g. These two benefits mimic the two-fold purpose of the see the right-hand side of the top part of Figure 1), and ISO 9001: 2000 standard, as stated in its scope section for other MS, such as the ones for occupational health (“demonstrate product conformance” and “enhance and safety (OHSMS) and corporate social responsibility customer satisfaction”). Other highly-reported benefits (CSRMS). Along this dimension of the M(E)SS, the main largely included “by-products” of such stated purpose challenges in research revolve around the discovery of (e.g. reduced customer complaints), while the benefits good methodologies for the implementation of the MSS unrelated to the standard (e.g. decreased employee themselves, as well as for the integration with the absenteeism) received a very low vetting rate. Hence, overarching MS. even though the use of e.g. ISO 9000 was found to be beneficial, for the MSS models to provide benefits Ascension outside of their limited scope, they must be expanded (Casadesus and Karapetrovic, 2005A). In difference to the augmentation-type MSS, which This point underlying the M(E)SS is not only cover only one specific element of an MS, the ascension- applicable in the static sense, but is even more type MSS relate to expansion of performance abilities of pronounced when time is taken into account. For the whole MS. This dimension is illustrated by the Y-axis example, the impact of specific MSS on companies, both in Figure 1. For example, a company can use an industry- in terms of inputs (e.g. costs) and outputs (e.g. gains), specific standard, such as ISO/TS 16949: 2003, to may actually decline over time, and newer MSS versions improve its ISO 9001: 2000 QMS, and then attempt to may be perceived as less beneficial than the older ones, as ascend towards excellence with the addition of ISO 9004: Casadesus and Karapetrovic (2005B) show for ISO 9000. 2000 - suggested concepts and techniques (e.g. see Boys Therefore, over time, MSS are subject to star-like et al., 2004), while other FS-MSS, for instance ISO transformations (e.g., when the ISO 9000 series changed 14004: 2004, provide similar facilities. from “quality assurance” to “quality management” in Unfortunately, and again due to the very nature of the 2000) and galaxy-like creations (e.g., when the ISO underlying MSS (e.g. see Dale (2004) for an explanation 14000 environmental series emerged in 1996). of the relationships between ISO 9000 and business Ultimately, the expansion of the M(E)SS from the excellence models), this M(E)SS dimension contains its minimum-requirement “seed” MSS (e.g. ISO 9001: 2000 share of difficulties. For instance, analyzing the or ISO 14001: 2004) is characterized with three spatial application of ISO 9004: 2000 in Canada, Boys et al. dimensions, which can be called “augmentation”, (2004) and Wilcock et al. (2004) conclude that not only is “ascension”, and “assimilation” (in homage to the this standard hardly used, but that conceptually, it seems popular “Star Trek”TM television series), and the fourth to suffer from the “lost lamb” syndrome, mistakenly temporal dimension (Figure 1). toying with the assimilation dimension of the M(E)SS. These problems, however, also bring about the need for Augmentation good research, specifically relating to the development of MSS frameworks along the ascension axis. Following the X-axis in Figure 1, an MS compliant with a seed MSS can be expanded to include additional Assimilation processes, which are developed according to different guidance standards and are subsequently built into the Expanding the initial FS-MS down the augmentation MS itself. For example, complaints-related processes and ascension axes creates parallel quality, modeled after the customer satisfaction complaint system environmental, and other similar function-specific (X-Y) (CSCS) standards (ISO 10001: 2006 for codes of plains, as shown in Figure 1. The third spatial dimension conduct, ISO 10002: 2004 for internal complaints (Z-axis in Figure 1) is obtained when different FS-MS are handling and ISO 10003: 2006 for external dispute assimilated into an IMS, thus essentially becoming its resolution) can be integrated into the quality MS (QMS) subsystems (e.g. see Karapetrovic, 2003). Such mirroring the ISO 9001: 2000 MSS (e.g. see Dee et al., assimilation may start with an existing subsystem (e.g. a 2004). Since the CSCS standards are still relatively QMS) forming the basis for the integration of other unknown, as two-thirds of this series has not even been subsystems (e.g. an EMS), or by putting the subsystems published yet, and as the only currently available MSS together using a shared or juxtaposed model. Regardless from the series (ISO 10002) is barely ten months old, a of the initial point, expansion along this axis may include section of this paper is devoted to an explanation of how a different number and functions of subsystems (e.g., they can be used in an IMS. QMS and EMS only, or QMS, OHSMS and CSRMS), Other augmentative function-specific management use diverse sequences (e.g. QMS first, then EMS and system standards (FS-MSS), such as ISO 10012: 2003 for OHSMS, or EMS first, then QMS), and may vary in the measurement systems, can be used for this purpose. The integration strength (e.g. subsystems may be left as same is true for an ISO 14001: 2004 environmental MS separate but harmonized, or be completely amalgamated) (EMS), which can be augmented with processes found in

20 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

M(E)SS ASSIMILATION AUGMENTATION ISO 14031

ISO 19011 ASCENSION TIME 4 0 0 4 ISO 10001, 10002, 10003 1 O IS ISO 10012 ISO 10015 3 0 4 0 0 0 0 9 9 O O IS IS 9 4 9 6 1 O IS

IMS

QMS

S M R S EMS C

QMS EMS

FIGURE 1: IMS and M(E)SS

.Time IMS As the status of each individual subsystem changes over time with respect to augmentation and ascension Concept (e.g. see the bottom part of Figure 1, where the EMS has been augmented, but has not ascended, while the QMS Within the M(E)SS space-time continuum, the has ascended without augmentation), and additional concept of an IMS refers to a unique set of assimilated, subsystems are assimilated, FS-MS and IMS are interdependent and function-specific subsystems, each themselves also dynamic in nature (Figure 1). established in accordance with one or more MSS, which Understanding the effects that time itself may have on share a collective pool of human, material, information, management systems (e.g. see Casadesus and infrastructure and financial resources to achieve both the Karapetrovic, 2005B), and conceptualizing flexible overall and function-specific goals (Beckmerhagen et al., models for dynamic integration along the three spatial 2003). Instead of developing FS-MS independently of dimensions of the M(E)SS, all provide opportunities for each other and largely at different times (top half of interesting research. Figure 2), namely when the corresponding FS-MSS become available or crucial for the company’s operation, the idea of an IMS is to plug the subsystems into an integrated framework, albeit still at different times as 21 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 dictated by MSS availability or necessity (bottom half of MS as the foundation for the corresponding IMS Figure 2). requirement, while additional MS requirements remain Building an IMS involves the integration of two main function-specific. This approach is fairly easy to aspects of the constituting subsystems, namely the implement, but is not adaptable to the introduction of new context (i.e. structure or the underlying MS model) and MS, and exhibits incomplete assimilation. For example, the content (i.e. substance or the MS requirements integrating the internal audit requirement of ISO 9001: covered), since both of these aspects differ from one MSS 2000 (clause 8.2.2), ISO 14001: 2004 (4.5.5) and OHSAS to another (Karapetrovic and Jonker, 2003). For example, 18001: 1999 (4.5.4) would leave risk assessment out of QMS standards are based on the so-called “process” consideration in setting the audit schedule for QMS and model, while EMS and OHSMS standards follow the EMS, as only the OHSMS standard specifically requires “PDCA” approach. In addition, FS-MSS may contain it. An integrated audit procedure would specify audit varying requirements even though they cover the same responsibilities, but not the methods used or MS element (e.g. internal audit), while their respective competencies required. MS can also be at differing levels of augmentation and Consensual Requirement: Here, the IMS element in ascension. question is derived from the universal characteristics of all FS-MS elements, while additional requirements found Context only in some particular subsystems are not fully assimilated. Although more comprehensive that its A single model for the IMS may be chosen using the “minimal requirement” counterpart, this notion provides following three alternatives (Figure 2): fairly similar advantages and disadvantages. For example, Initial Model: In this strategy, the model of the seed although consideration of importance to quality, FS-MS forms the underlying IMS framework. For environmental and safety aspects and impacts would be example, the “process” or the “PDCA” model would be added in determining the audit schedule, risk assessment used if an ISO 9001-complaint QMS or the ISO 14001- would still be excluded, and the audit procedure would registered EMS was established first, respectively. This not contain provisions on auditor competence. strategy is probably the easiest one to use, capitalizing Maximum Requirement: In this approach, the most from the already-operating and entrenched MS, but does comprehensive requirement of any function-specific MSS not benefit from any newly-added MSS models, and provides the basis for the IMS, therefore demanding total requires forceful adaptation of any new MS to the assimilation and providing the most comprehensive and existing model. flexible strategy. For instance, the assessment of risk with Combined Model: Companies may also put together respect to product quality would be taken into account the models of the constituting MS into a single founding when scheduling the audit, and an integrated audit model. For instance, “process” and “PDCA” approaches procedure would address how auditor competence is can be integrated to form an overarching model for an ensured, regardless of the specific subsystem covered. IMS with QMS, EMS and OHSMS components. This approach is advocated by the MSS themselves (e.g. see the introduction sections of ISO 9001: 2000 and ISO 14001: 2004), and makes sense in the short term with the CSCS implementation of the most popular MSS. However, problems would occur if a new FS-MSS, such as the CSRMS one, is based on a model that does not Concept necessarily conform to the combined approach, or if a The Customer Satisfaction Complaint System or constituting MSS, e.g. ISO 9001, is revised to include a (CS)2 series contains three MSS, namely ISO 10001 completely different model. “Guidelines for Codes of Conduct”, ISO 10002 Accommodating Model: A model that adapts itself to “Guidelines for Complaints Handling in Organizations”, the existing and future MS approaches is implemented in and ISO 10003 “Guidelines for Dispute Resolution this alternative. Such “generic” models have been External to Organizations”. The series is being developed reported in literature (e.g. see Karapetrovic and Willborn, within the Technical Committee (TC) for Quality 1998; Wilkinson and Dale, 2002). This strategy is the Assurance and Management of the International most flexible one, since even a company’s own model Organization for Standardization (ISO), specifically by can be used, without forcing conformance to the specific the Subcommittee (SC) on Supporting Technologies FS-MSS models. Such models can also be easily adapted (ISO/TC176/SC3). Upon an initiative by the ISO to changes in the current MSS and additions of new MSS, Consumer Policy Committee (COPOLCO), the actual and should be able to accommodate all three M(E)SS work started in May 2001, with the completion expected dimensions. Problems in reaching consensus in the by the end of 2006. ISO 10002 is currently the only company regarding the actual model to be selected publicly-available standard in the series, appearing in represent a potential downside. July 2004, while ISO 10001 and 10003 will be published simultaneously in 2006. Content Three approaches are also available when the assimilation of content is considered (Figure 2): Minimum Requirement: This strategy involves the application of the “minimum” requirement shared by all 22 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

… ISO 9001 (t1) ISO 14001 (t2) ISO XXXXX (tn)

QMS EMS XMS …

ORGANIZATIONAL SYSTEM

ISO 9001 (t1) ISO 14001 (t2) … ISO XXXXX (tn)

INTEGRATED MANAGEMENT SYSTEM

Initial Model Combined Model Accommodating Model

“PROCESS” “PROCESS”+ “SYSTEM” ”PDCA”

[ISO 9001] [ISO 9001/14001] [IMS]

OBJECTIVES PROCESSES … RESOURCES

PLANNING OPERATION … IMPROVEMENT

… … …

Minimum Requirement Consensual Requirement Maximum Requirement

ORGANIZATIONAL SYSTEM

FIGURE 2: IMS Context and Content

Although they can be used independently, the standard, these MSS will have identical structures, greatly standards are logically interconnected and numbered assisting in their mutual compatibility and in further following the life of a product quality-related complaint, integration (Dee et al., 2004). Therefore, each CSCS from the establishment of a structure to handle the standard contains sections: 4 “Guiding Principles”; 5 complaint with a quality guarantee in the form of a code “Framework”; 6 “Planning / Design”; 7 ”Operation / (ISO 10001), to the resolution of the complaint internally Implementation”, and “8 Maintenance and (ISO 10002) or externally (ISO 10003) (Dee et al., 2004; Improvement”. Dee et al. (2004) provide further details Hughes and Karapetrovic, 2005). Apart from the ability on the CSCS series. to provide the foundation for creating a system of As with any other management system integration, sequential complaint-related operational processes in a both the elements (i.e. content) and the specific company, which are covered in section 7 of each relationships among the elements (i.e. context), of the MS

23 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 established according to the CSCS series of standards standards is fully compatible with both ISO 9001 (e.g. should be assimilated. Figure 3 illustrates some “framework” in ISO 10002 is analogous to “management integration alternatives, with ISO 9001, ISO 14001, and responsibility” in ISO 9001) and ISO 14001 (e.g. ISO 10001, 2, 3 providing an example of the scope of “maintenance and improvement” in ISO 10002 is such assimilation. analogous to “preventive and corrective action” and “management review” in ISO 14001), CSCS can be put Context into a QMS, EMS or even an IMS (using an accommodating model, such as the systems one from Observed from the bottom up, Figure 3 shows two Karapetrovic and Willborn, 1998). integration paths for CSCS processes, namely the code In the second alternative, a company would first process (ISO 10001), the complaint-handling process establish separate processes according to each of the (ISO 10002) and the dispute resolution process (ISO CSCS standards. Then, several sub-sequences are 10003). In the first alternative, process elements which possible, for example, CSCS to QMS to IMS, or CSCS to deal with individual outputs (code in ISO 10001, EMS to IMS, or CSCS directly to IMS, or even skipping complaint in ISO 10002, dispute in ISO 10003) can be over the CSCS amalgamation altogether and integrating, connected horizontally first, and then vertically into an for example, the complaints-handling process only, with a overarching CSCS framework. Subsequently, as Dee at QMS and IMS. al. (2004) point out, since the structure of the CSCS

GOALS PROCESSES RESOURCES

Determine Plan & Design Acquire

QMS, EMS, CSCS QMS, EMS, CSCS QMS, EMS, CSCS Community Customer

Evaluate Implement Deploy Complainant Provider QMS, EMS, CSCS QMS, EMS, CSCS QMS, EMS, CSCS

INTEGRATED MANAGEMENT SYSTEM (QMS, EMS, CSCS)

Management Policy Responsibility Management CSCS Measurement, Review CSCS Resource Analysis & CSCS Management Improvement Planning

CSCS Product CSCS CSCS Checking & Realization Corrective Action Im plem entation CSCS CSCS & Operation CSCS QMS (ISO 9001) EMS (ISO 14001)

Framework

Planning & Maintenance & Design Improvement Implementation / Operation

Guiding Principles

CUSTOMER SATISFACTION COMPLAINTS SYSTEM

Complainant Complainant Framework

Planning & Maintenance & DISPUTE CODE PROCESS Design Improvement RESOLUTION (ISO 10001) Operation PROCESS (ISO 10003)

Guiding Principles Provider COMPLAINT – HANDLING PROCESS (ISO 10002)

Communicate Receive Acknowledge Assess

DISPUTE CODE Track Complaint RESOLUTION IMLEMENTATION OPERATION (ISO 10001, (ISO 10003, Section 7) Close Decide Respond Investigate Section 7)

COMPLAINT – HANDLING OPERATION (ISO 10002, Section 7)

FIGURE 3: IMS and CSCS

24 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

/6/ Dee, B., Karapetrovic, S., Webb, K. (2004), “As Easy Content As 10001,2,3”, Quality Progress, Vol. 36, No. 6, pp. 41-48 /7/ Hughes, S., Karapetrovic, S. (2005), “ISO 9001 and Since CSCS standards have been drafted to maintain ISO 10002 in an Electrical Utility”, Proceedings of the 10th compatibility with other ISO MSS, specifically with ISO International Conference on ISO 9000 and TQM, Shanghai, 9001: 2000, elements of complaint-related MS can be China, pp. amalgamated into respective elements of QMS or IMS /8/ Karapetrovic, S. (2003), “Musings on Integrated without much trouble. For instance, the measurement, Management Systems”, Measuring Business Excellence, Vol. 7, analysis and improvement processes in an ISO 9001: No. 1, pp. 4-13 2000 – based QMS can be augmented with complaint- /9/ Karapetrovic, S., Jonker, J. (2003), “Integration of handling operations from ISO 10002: 2004, including e.g. Standardized Management Systems: Searching for a Recipe and audits, corrective actions, and others. Ingredients”, Total Quality Management and Business Excellence, Vol. 14, No. 4, pp. 451-459 /10/ Karapetrovic, S., Willborn, W. (1998), “The Systems View for the Clarification of Quality Vocabulary”, International Journal of Quality and Reliability Management, Vol. 15, No. 1, pp. 99-120 CONCLUSION /11/ Wilcock, A., Karapetrovic, S., Boys, K., Piche, P. (2004), “Towards Business Excellence: The Case of ISO 9004 These are exciting times in the realm of management in Canada”, Proceedings of the 9th International Conference on system standardization and integration. Numerous ISO 9000 and TQM, Bangkok, Thailand, pp. 50-58 ongoing international initiatives to develop solutions for /12/ Wilkinson, G., Dale, B.G. (2002), “Integrated the integration of internationally-standardized function- Management Systems; A Model Based on a Total Quality specific management systems are a good indicator of Approach”, Managing Service Quality, Vol. 11, No. 5, pp. 318- such excitement. For example, a working group of the 330 ISO Technical Management Board is currently preparing a “Handbook for the Integrated Use of Management System Standards”, which should provide some good methodologies and examples of integration. This paper attempted to provide some insights into this realm, by discussing integration from the perspective of management system standards and focusing specifically on a new series of these standards related to complaints. While it can be expected that future research will be devoted to the development of models for and analyzing the impacts of management system standards themselves, research on the integration of management systems should be concentrated on generating new methodologies, rather than models that there are plenty of already in existence.

REFERENCES

/1/ Beckmerhagen, I.A., Berg, H.P., Karapetrovic, S.V., Willborn, W.O. (2003), “Integration of Management Systems: Focus on Safety in the Nuclear Industry”, International Journal of Quality and Reliability Management, Vol. 20, No. 2, pp. 210- 228 /2/ Boys, K., Karapetrovic, S., Wilcock, A. (2004), “Is ISO 9004 a Path to Business Excellence? Opinion of Canadian Standards Experts”, International Journal of Quality and Reliability Management, Vol. 21, No. 8, pp. 841-860 /3/ Casadesus, M., Karapetrovic, S. (2005A), “The Erosion of ISO 9000 Benefits: A Temporal Study”, International Journal of Quality and Reliability Management, Vol. 22, No. 2, pp. 120-136 /4/ Casadesus, M., Karapetrovic, S. (2005B), “An Empirical Study of the Benefits and Costs of ISO 9001: 2000 Compared to ISO 9001/2/3: 1994”, Total Quality Management and Business Excellence, Vol. 16, No. 1, pp. 105-120 /5/ Dale, B.G. (2004), Managing Quality, Blackwell Publishers, London, UK, 4th edition

25

Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

A CONTRIBUTION TO THE DIGITAL QUALITY CONCEPT RESEARCH Keynote paper Professor Vidosav D. MAJSTOROVIĆ, Dr.Sci., Mech. Eng. Mechanical Engineering Faculty, Belgrade, Serbia

Summary: The development and research of new manufacturing systems’ generation is today based on new paradigms in education, research and implementation. Each of the above entities has its own detailed structure related to the development and application of the digital factory concept. This concept integrates digital product, digital planning, digital workshop and real factory. This paper analyzes and reviews this concept as well as some elements of obtained research results related to digital quality, as a concept for QM in these factories. Key words: Digital factory, Digital quality, Research.

1. IN LIEU OF AN INTRODUCTION lowest price, expect the greatest values for their money. Therefore, the result of all this is the pressure on Increasingly stronger demands for products with producers to improve the quality of their products and shorter innovation cycles are placed on to-day’s services, and digital factory is today’s answer to all these generations of manufacturing systems, thus leading to the challenges /8/. decrease of life cycle and diversity in the number of the variants of products. IT technologies and rapid prototype development technologies represent the approaches that are effectively solving the above demands with the aid of 2. DIGITAL FACTORY CONCEPT digital factory concept. EU industry has the most important role in the development, particularly in the The effective management of information about domain of increasing the European competitiveness at the products and manufacturing resources are key elements global market. Annual gross product of the European of the digital factory, that may be defined on three levels: industry today amounts to 4,000 M euros, with direct (i) definition of the total geometric model of the factory employment of 40,000 workers and the newly acquired (product, process, workshop), assuming that digital value amounts to 32%. Sustainability and advancement factory is a prototype of the model for ergonomic layout of the existing role of the industry in the economic, social planning. The changes at any of the levels may be easily and technological development implies additional and quickly effected, which otherwise represents a great investments into I&R high-tech and science-based limitation in the existing factories, (ii) at this – the second industries. EC has therefore defined the research and level, static model is transformed into a dynamic one with technological development strategy on the basis of the the aid of simulation tools and techniques. In this way advancements in the innovation chain that includes: the the virtual factory is obtained that may be optimized institutions (of the EU member states), universities, according to different criteria, and the most frequently research institutes and centres, companies. As a result of according to unit costs. Virtual configuration may be the already applied industrial concept of digital factory frequently changed in this fashion, for example – once a has been developed out of these demands /14/. week, (iii) the third level is the fusion of virtual and real The economy of Europe is faced today with new model and a real factory. The new generation of real economic environments, globalization and technological factories is exclusively built with the use of simulation development processes. The need is therefore felt for models. After they are put into operation, the basic data new management approaches, and public administration are built into a simulation digital factory model. Digital is expected to create a new environment that should factory integrates the generation, memorizing and support sustainable development of the EU society. The delivery of all relevant data about the product and process strategic goal of the economy is the advancement and in its life cycle – from the earliest planning to the mass creation of the level of competitiveness that will provide production. Digital factory is much more than a 3D higher standard to the EU citizens. On the other hand, virtual factory, which with the simulation process EU has its own vision of leadership and world connects product/process through management with the perspective in political, economic, technological and data about the product and organization’s management cultural, i.e., civilization values for all its states, nations and the flow of material. Digital factory concept supports and citizens. The possibilities and challenges of the cooperative engineering model which developed from the modern information society play a key role in the creation simultaneous engineering model. This model resulted, of new forms of partnership between companies, social however, from computer aided automation of engineering community, trade unions, consumers, industry and activities - CAx technology. organizations in private, mixed or state ownership. The The intensive development of information- demands of buyers are constantly increasing. They do communication technologies has opened new possibilities not accept inferior products, demand the best at the for automation of factories. Different software tools

27 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 support engineering activities in production engineering, represent the basis for the specification of IT system, (ii) production planning and the production itself. The vision the integration of data between digital product and digital of the digital factory is realized in automobile industry factory are the main factor for successful realization of with the use of integrated engineering production this concept, and (iii) the integration of data is limited by planning process supported by coherent methodology and compatibility and possibilities of the IT system, computer technology. The automobile industry is the particularly with their interfaces. The directions for leader in the application of digital factory concept, where further improvement of digital factory performances in total integration of product and process design is applied this field are: (i) decrease in the number of different from the earliest planning stage to the end of the systems and models of data, (ii) independent and separate production. In this way a number of product variants models for data management, (iii) independent individual (mass manufacture according to unit buyer’s demands) is interfaces, and (iv) independent systems for redundant drastically increased, with a great decrease in the data management. In the field of design and planning the development time, which means: (i) shortening of decrease of applied number of different systems indicates development cycle, design, production and testing time, the selection of CAx/EDM system on the same platform. (ii) quick results of product characteristics, and (iii) fast In the manufacture, however, this system is not possible and simple exchange of data in cooperative engineering because of the great number of specialized systems for model. In this way the integrated approach is applied in tool management, NC programming, quality practice in the field of production engineering. management, etc. A possible solution in this field The virtual tools based on IT technologies support the (CAx/EDM) is the integration of systems on an complete process of product design and process independent data model, describing independent models development in the digital factory. Central focus in (geometric, functional, logical) between products, product development is directed to the (i) analysis of processes, resources. These models are also linked materials and structural analysis, (ii) technological through different systems. Such approach enables a feasibility studies, (iii) development of tools and uniform approach to different data structure memorized accessories and (iv) matching and management of in different independent systems. The integration of tolerances. In the development of technological process specialized IT-tools with their own data management the attention is directed to the (i) determination of models and their performances represents, however, an machining procedure, (ii) definition of assembly imperative for digital factory, where integrated model for procedures, (iii) tool management, (iv) factory layout, (v) the design, planning and manufacture is realized and ergonomic simulation, and (v) effective simulation of the used. flow of materials. This concept is supported by the The greatest number of technological processes may following virtual tools: I-DEAS, Unigraphs, CATIA, be today modeled and simulated. It is also customary, ROBCAD, Diemaner, Pam-stamp, Autoform, Viking, however, to realize the simulation of technological Simple, Microstation/Autocad, eM-Planner, and others. processes as a separate model. A special challenge for The application of digital factory concept made the the realization of a digital factory is the integration of greatest progress in automobile industry, where the simulation results in different technological processes in following effects are brought about: (i) impact on the design and planning. With simulation tools a bridge products (drastic increase in variants of different basis must be established between the description of local models; increase in product’s quality; costs decrease), processes and global process characteristics, which and (ii) effects on the process (shortening of time to represents a chain of local processes. The basis for this appear at the market, development and manufacturing simulation consists from data arriving from product time; continuous change in process management; development (3D geometry and tolerances, material, flexible high performance manufacture). From the etc.), and production planning (clamping tools and practical point of view the basic steps in the application accessories, technological operations, etc.). The of digital factory concept include: (i) CAx/EDM integration of product development and simulation of integration, (ii) integration of simulation results into the technological processes is achieved in four steps: design and planning , (iii) management with 3D experiment, process optimization, generation of the tolerances and process chains, and (iv) achievement of optimum process and transfer to the production. The feedback for quality management. These elements have tolerances must satisfy the technical and economic made possible: (i) the integration of the user into the requirements. However, the determination of tolerances early stages of product and process development (QFD is an interdisciplinary task carried out by teams from method), (ii) constant harmonization and adjustment of process chains consisting of the designer, production new user’s demands related to new products, and IT planner, and quality planner. The iterative planning tools, and (iii) the cooperation with the internal and process of tolerances includes: 3D modeling of external deliverers and their optimization. Digital tolerances, their visualization and process validation. The factory connects digital product, digital planning, digital concept of determining tolerances includes the definition workshop and real factory. In further text the design and of essential information and resources for the design and planning, simulation and quality management aspects in product and process planning. In the analysis and digital factory are analyzed. synthesis of this concept a large number of developed IT When considering CAx/EDM as a crucial field for simulation tools is used. The basic goal is to define the the functioning of digital factory, the following elements optimum between necessary tolerances and product’s stand out as especially important: (i) the processes in characteristics. The starting element for the simulation of digital factory must be defined very well because they tolerances is the model containing 3D geometries and

28 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 tolerances, clamping and auxiliary accessories, real phase of production planning. By comparing them with machining operations, as well as a real tolerance the targeted once from the previous phase, the economic distribution, where the tools based on stochastic methods effectiveness for the product is reached. are used in this field. Thus, for instance, the variations in Product design. This phase is crucial for the manner of clamping or the machining operation may technological costs. The analysis of product design be compensated by measurement chain method process in the digital factory concept is made with the aid (compensation member) without changes in tolerances. of different IT tools (DFM, DFI,…), thus taking into In view of this, the problem of tolerances is solved in consideration, in this phase, the effects of later phases of product design and process planning stages, thus its development (production, inspection, assembly,…). eliminating additional costs. In this way the integrated Process planning. Together with product design, its model for solving of tolerances is defined. technological analysis is made (manufacture, control- The last step/element in digital factory that provides inspection, assembly). These processes are defined as constant harmonization and synchronization between it flow diagram with resources essential for these processes. (digital factory) and a real factory is feedback-based Manufacturing concept design. The focus in the quality management. In the previous stage the tolerances creation of manufacturing concept is the definition of have been defined and solved, and in this stage they are production management model and the tasks with realized in the production, with the possibility of their specified requirements. This model must be compatible constant improvement. The first step in the realization of with factory concept. If the digital factory model is feedback concept is the prognosis and diagnosis of data developed without these elements, the discontinuity will for conformity quality management on the basis of goals, arise between the virtual product model and the factory. and applied required function methods. The precondition Planning of the factory and logistic support. In the for continuous use of data on quality is explicit use of planning of the factory and its logistic support, the focus information about tolerances, as well as measured values is put on the development of different layout variants for of these tolerances after machining process. These data the designed production model. 3D visualization and (planned and measured tolerances) are generated into a multi-medial layout presentation enables the planning of separate data base producing independent product’s the flow of materials as well as logistic planning. quality model that is a part of the integrated product and The design of working conditions. This element process model. The basic metrological system used in relates to detailed definition of technical and spatial this concept is CMM that enables: (i) the use of parameters of technological and assembly working posts. standardized procedures for generation of metrological It includes the designing of work methods, ergonomic data, (ii) robust generation of data on quality out of the aspects of working posts and installations, integral layout process, and (iii) explicit assessment of measurement of the work process as well as the organization of work. results. In this way, the feedback concept is fully realized The concept connects the worker, technology, through the process data and planned data on quality information and organization from the economic aspect (tolerances). In this fashion are also created the data base with optimum conditions. and knowledge base that includes experimental and Time management. Standardized methodological expert knowledge in this field. Thus the integrated procedures require the use of data about time resources quality model of the digital product in the digital factory for different phases of the design, planning and product’s is obtained. manufacture. They provide the basis for the management of technological and manufacturing cycles, improving productivity in technological fields. Simulation. 3D simulation of the flow of materials is 3. DIGITAL FACTORY AND DIGITAL carried out by variation of a large number of factors (the QUALITY MODEL number, management and size of warehouses – tools, materials,…, clamping working piece strategy, Digital factory is realized as an engineering concept in disorders/failures on the machine, etc. This is necessary practice, especially in automobile industry. It passes in order to get the dynamical machine model as close as from the experimental-research workshop phase into the possible to the reality. phase of diffusion towards the industry. This is why Virtual factory. 3D visualization and multi-medial today we may speak about the project stages of its simulation of the machine and equipment with their realization for the application in practice /7-15/. dynamic environment in interaction with the employees The definition of project requirements. Digital factory and the environment gives a virtual model of the digital project begins with the analysis and assessment of factory. In this way, negative effects of possible mistakes conditions for the product and its production. in real environment are avoided. All the above manufacture The initial parameters include: targeted mentioned phases are supported by PARTIAL and costs, expected sale, possible prices (all for the product), integral model of the product and factory, as a basic as well as essential resources for the planned product. virtual model of the digital factory. Analysis of functions. At the beginning of Digital quality. Quality management in digital engineering analysis of the product it is necessary to factory, of its partners and suppliers is carried out by the describe the manufacturing functions. With this in mind, use of current approaches in this field. The basis for the a detailed analysis of costs for the product is to be made development and application of this concept is digital first. In this way planned costs are defined at the earliest quality model with several dimensions, discussed in the following text. The quality as an economic-commercial

29 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 concept represents the characteristics of the product that this type of quality. The second type is the design for is planned and produced for the user. The manufacturer quality, known as the approach to quality of should provide the quality demanded by the buyer, and performances. This approach is using professional the buyer demands what he desires. This is why the knowledge and experience of experts, generated for manufacturer is constantly learning from buyers’ certain fields. Current characteristics of product’s or demands, and in order to achieve the required quality he service quality are compared with theoretical or best uses different quality engineering techniques. The practice indicators. In this case, therefore, quality supplier is not satisfying only buyer’s demands through represents the relationship between the actual designed the quality, but also other organization’s interest groups, solution and the level of performances achieved in the although these may be in conflict. The most important production. The quality is then obtained as a design product’s quality characteristics are recognized and excellence, dependable also on the scientific and accepted by the buyer at its delivery. At the open market, technical progress and application of computer the role of public administration is to define the minimum technology. The conformity and quality performances of the demands and to control whether they are are compared with achieved production and design satisfactory as to the health of the consumers and the elements. The improvement of product/service protection of the environment. Unexpected quality competence from the buyer’s view represents the third properties of the product, obtained after its delivery, give type, which is called buyer’s quality. The elements of to the quality immanent (internal) characteristics termed product/service quality are generated by different as transcendental (difficult to explain) characteristics. It marketing methods, such as the information about may be therefore said for quality that it is the feeling buyer’s satisfaction, focused on specific groups. They (expectation) of full satisfaction of expectations, and an are obtained through the comparison of buyer’s independent entity of a product or service. In the delivery expectations and achieved product/service quality of a product/service, the transcendental quality is not the characteristics. In this approach the buyer has a twofold subject of defined requirements (values). The term ex influence in relation to the producer MANUFACTUER, post quality denotes the quality displayed during the use. and these include expected performances and the price. Contrary to this, the term ex ante denotes the expectations This approach in the definition of product’s quality is of quality indicators before the delivery. The majority of particularly obvious in automobile industry (Mercedes, markets today function according to the principle of ex VW). When buyer’s quality is significantly improved, ante quality. The world leaders in quality, particularly in the quality of living is also improved, which represents the field of industrial products, are taking over the fourth type of quality. In this case the expectations of unexpected innovations that in the ex ante phase generate the buyer are fully satisfied within the chain: groups of the feeling of an attractive quality, or the quality above interest – system (organization) – quality of the expectations. The elements for such innovations are environment. This is achieved through the integration of obtained through generated experience, knowledge and the management of the system for quality, living learning from ex post quality, that may be multiplied as a environment, safety of the employees and the protection result of the best practice. Together with the notion of of their health. This concept is today certified as an transcendental quality the following question may be also integrated management system (IMS) including: QMS defined - Could the principles of quality management, (ISO 9001:2000) – EMS (ISO 14001) – OHSMS the technique and methods of quality engineering, be (OHSAS 18001). The above four concepts use different successfully applied in the development of this concept? sources of information and methodologies for quality In the market economy, the buyers are free to state what management, having primarily in view the facts that they wish or expect from the product. Therefore the represent the basis of all approaches, that is: (i) provision producer translates these questions (wishes) into a of quality in the manufacture and the design (designing practical form – defined requirements that must be for production/quality, QFD methods for translation of satisfied at the delivery. Starting from this approach, four buyer’s demands into project specifications), (ii) types of ideal quality may be defined, which represent a excellence models oriented towards the satisfaction of the source of information for such requirements, that is, the requirements of organization’s groups of interest, (iii) answer to the question – How can the manufacturer know systems for the protection of the environment, and (iv) what is the quality of a product/service for the buyer? social quality or the quality of the society. The first type is the quality known as a flawless manufacture, that is, conformity quality, where the information about quality requirements are generated from technical drawings, contracts and specification of 4. IN LIEU OF THE CONCLUSION services. The inspection, quality control, audit, re- testing, corrective and preventive measures, as well as In view of all of the above, we may conclude that the bad quality costs are the methods for the achievement of concept of the digital factory is based on the management conformity. In big series and mass production the of data and resources about the product, the planning and following is used: statistical theory of variations, the manufacture, that is, on all processes that occur in problem solving techniques, the design based on these chains. In this entire concept the module of experiments and standardizations, as well as very tolerances is the primary one because it enables the successful tools for the achievement of harmonization. feedback-based management of the conformity quality. Quality management systems (QMS), that is ISO Current research in this field relates to: (i) how to 9000:2000 represents the latest approach in this field for connect and manage as effectively as possible different 30 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 clusters of information about the elements of digital product’s model and their links, (ii) how to manage an increasing number of product variants in their digital version, (iii) how to integrate the partners and suppliers into the entire digital factory concept with all necessary information about the digital product, and (iv) the research and the development of the new generation of engineering application software for product design, process planning and production control, particularly for intelligent quality in digital environment.

REFERENCES

/1/. UNICE, Stimulating Creativity and Innovation in Europe, The UNICE Benchmarking Report 2000, Bruxelles, 2000. /2/. Jovane, F., The role of Research in Manufacturing in EU, ITIA-CNR, Institute of Industrial Technologies and Automation - National Research Council of ITALY, Milano, 2004 /3/. EC, Workshop "The Machine-Tool sector and the FP6", Leuven, 2002. /4/. Jovane F., “Research based Evolution of the Man- Industry Value Chain”, ITIA series, MRI, 2002. /5/. EC - DG Research, “The Future of Manufacturing in Europe 2015-2020: The Challenge for Sustainable Development (FuTMan)”, Bruxelles, 2002. /6/. EC, “Report on the analysis of Expressions of Interest”, http://www.cordis.lu/fp6/eoi-analysis.htm, 2002. /7/. EC, RTD-Comdocuments Database of Cordis, http://www.cordis.lu/en/src/d_001_en.htm. /8/. Wagner, T., The Digital Factory, Bosch GmbH, Stuttgart, 2004. /9/. Westkamper, E., Smart Factories – Digital Factory as System of the Future, Institute of Industrial Manufacturing and Management, Stuttgart, 2004. /10/. Bar, T., Steps Towards the Digital Factory, DC Resarch and Technology Center, Ulm, 2004. /11/. Zhao, J., A Consistent Manufacturing Data Model to Support Virtual Enetrprises, International Journal ‘’Agile Manufacturing Systems’’, Vol.8, pp.150-158, 2005. /12/. Davidrajuh, R., An Autonomus Data Collection System for Virtual Manufacturing Systems, International Journal ‘’Agile Manufacturing Systems’’, Vol.7, pp.7-15, 2004. /13/. Maropoulos, P., Digital Enterprise Technology, International Journal ''Computer Integrated Manufacturing'', Vol.16, pp.468-479, 2004. /14/. Gao., J., et all, Application of Product Data Management Technologies for Enterprise Integration, International Journal ''Computer Integrated Manufacturing'', Vol.16, pp.491-505, 2004. /15/. Peggs, G., Virtual Technologies for Advanced Manufacturing Metrology, International Journal ''Computer Integrated Manufacturing'', Vol.16, pp.485-492, 2004. /16/. Majstorović, V., Digital Factory – Fiction, Reality or Future, XXX JUPITER Conference, pp. 12-18, Belgrade, 2004.

31

Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

ORGANISATIONAL SUSTAINABILITY MANAGEMENT THROUGH MINIMISED BUSINESS EXCELLENCE MODELS

Keynote paper Rickard Garvare1) , Raine Isaksson2) 1) Luleå University of Technology, Luleå, Sweden, 2) Gotland University, Visby, Sweden.

Summary: Unsustained development is no real development. But for most organisations development is a necessity for survival. Therefore sustainability should be of interest to all organisations. To enable participation by everyone, including small organisations, methodologies and tools for sustainable development has to be straightforward and simple. Based on principles of Business Excellence and process management this paper presents a conceptual framework for minimised models aimed at assessing and improving organisational sustainability in terms of economic prosperity, environmental protection and ethics. An indicative application in the form of a case study is also provided. Keywords: Management, Excellence, Business

INTRODUCTION which could delay the start up of organisational reporting. Consequently the GRI is encouraging organisations to The drivers for sustainability could for instance be commence measurements with a reduced number of moral, legal, political or commercial. Irrespective of indicators, GRI (2002). Our study emphasises the reason it seems justified to consider how organizations importance of keeping measurement systems straight- could improve their performance in terms of forward and simple enough to be understood by ordinary sustainability, see Zairi & Peters (2002), Aaronson people and not only by experts. (2003) and Bryane (2003). Issues of Sustainable It seems clear that there should be synergies in Development (SD) are essential to all types of integrating Total Quality Management (TQM) and organisations, Edgeman & Hensler (2001). In our view, Sustainable Development (SD). Hellsten & Klefsjö the importance of the subject suggests that measurements (2000) describe TQM as a management system consisting of SD should to a larger extent be included in of values, methodologies and tools. When going from management systems such as ISO 9000 and Business TQM into a combined TQM-SD structure, the values, Excellence Models (BEM) such as the European Quality methodologies and tools have to be reviewed, but as Award and the Malcolm Baldrige National Quality discussed by Isaksson (2004) the same approach should Award, see Garvare & Isaksson (2001). Quality be applicable. Recognizing the combination of TQM and management systems are increasingly being integrated SD as a management system provides the freedom to with business management systems that also include define which elements are included and to clearly state health, safety and environmental issues, Wilkinson & the content of the approach in a particular case. From this Dale (1999), Karapetrovic (2002), Karapetrovic & Jonker perspective the process view, being a core value of TQM, (2003). This emphasises the necessity of finding agreed and also emphasised in ISO 9000 (CEN, 2000), should indicators for all dimensions of SD. preferably be included in models measuring SD- Several authors describe the reporting of performance. Process orientation play an important part organizational sustainability, using the Triple Bottom in many methodologies for improvement, such as self- Line (TBL) that divides results in economic, assessment based on BEM criteria, and Six Sigma environmental and social performance; see Elkington improvement, see Magnusson et al. (2003). (1999), and Töpfer (2000). One of the best known According to GRI (2002) the GRI guidelines do not initiatives for measuring organisational TBL are the seek to divide performance indicators into types based on guidelines of the Global Reporting Initiative (GRI, 2002) the content or nature of the indicator, e.g., based on mentioned by the United Nations Environment policy, input/output or impact. Furthermore, the Programme (UNEP) and the World Business guidelines do not emphasise the value of focusing on Commission for Sustainable Development. Central processes. There could be a risk of misinterpretation principles for the GRI reporting are transparency, when speaking about a bottom line and then including inclusiveness and auditability. Companies are required to policy and input in it. Our contention is that quality report on the output and outcomes of stakeholder related indicators should form part of an indicator system consultations. The GRI guidelines classify indicators in based on the Triple Bottom Line. Currently this is not the the economic, environmental and social dimensions. case with the GRI-guidelines. In the GRI Economic These indicators are either core indicators or additional dimension the proposed indicators are measures of indicators. For reporting in accordance with the GRI monetary flow, such as net sales and the distribution of guidelines all core indicators need to be included. Out of costs. This would indicate that an organisation could the total of about a hundred indicators about half are core. claim economic sustainability when having a surplus, The number of indicators is relatively high, something irrespective of the efficiency of the operations, cost of

33 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 poor quality or what value is generated for the customers. Values, methodologies and tools According to Adams (2004) a radical overhaul of corporate governance structures is required in order to A BEM relies on its core values. When combining close the gap between official TBL reporting and actual TQM and Sustainable Development (SD) the values SD performance of companies. Edgeman & Hensler should reflect both areas. Garvare & Isaksson (2001) (2004) conclude that “commitment is required regarding propose five core values for SD being sustainable stewardship of economic, environmental and human stakeholder balance, learning excellence, process resources – the conscious choice of service over self- performance excellence, stakeholdercracy and interest with a concept of profit embracing the notion of transparency. Even though TQM clearly acknowledges residue – that which remains after all obligations are the concept of interested parties, focus is still largely fulfilled.” placed on strong customer orientation. In TQM-SD there Hendricks & Singhal (1997) conclude that the use of should be a clear shift towards balancing the needs of Business Excellence Models (BEM) is often a successful different stakeholders. Edgeman & Hensler (2004) way to work with TQM. Traditional self assessment underline the importance of reaching a perspective on based on BEM is a demanding exercise requiring stakeholder needs that extends beyond a limited zero-sum considerable resources. This could make the approach game and into a win-win scenario. The process focus is complicated for smaller organisations, see Eriksson & an important value supporting many methodologies and Garvare (2005). Finding an easier way of starting the tools that should be as effective in customer focused as work for sustainability should be of interest for many stakeholder focused improvement work. Several other organisations. values could of course also be discussed but this is left The purpose of this paper is to present a framework for further research. One important contribution of SD is for minimised models of assessing organisational the emphasis on transparency, a value which aids sustainability in terms of the Triple Bottom Line. The focusing on methodologies and tools that counteract results of such an assessment should indicate the corruption and distortion of market mechanisms. organisational improvement potential, in terms of both Corruption has effect on all dimensions of the Triple process performance and resource utilisation. Our Bottom Line (TBL). The Environmental Sustainability intentions with this paper are to: Index (ESI), that ranks national sustainability performance, mentions corruptions as one of the • review the Triple Bottom Line content indicators with the strongest correlation to the entire • review the values of an integrated TQM-SD system in index, ESI (2002). order to create a basis for realising expected synergies Central values for the discussion in this paper are the use a generic organisational process model to identify • focus on stakeholder needs, process orientation and TBL-performance and enablers as self assessment transparency. These values support the use of process criteria management and measurements based on the TBL as well • ensure that the presented framework for assessment as BEM. models could be customised for the specific organisation in such a way that the most relevant Process models criteria will be prioritised. A specific process chart, based on the common By means of a grounded approach based on division of organisational processes into management, Innovation Action Research, see Kaplan (1998), we have operative and support processes, has been developed as a been testing the initial ideas in a case study. The aim of tool for process management, see Figure 1. this case study has been to analyse the feasibility of the model framework and also to identify focus areas for future research. This early presentation at such a conceptual stage of the research has the purpose of inviting others to take part in the discussion.

FRAME OF REFERENCE

Total Quality Management (TQM), process models, the Triple Bottom Line as described by the Global Reporting Initiative, Business Excellence Models (BEM), and the idea of the vital few, presented by Joseph Juran, have all contributed to the frame of reference used for this study.

34 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Process or organizational mission: Converting input to output/to convert input to output (using verb with ing-form or to + verb + noun) Management Processes Management process 1 Management process n

Operative Processes Sub-process A1 Sub-process An

Input Output Sub-process Z1 Sub-process Zk

Support Processes

Support process 1 Support process n

Resources (enablers): Management, Man, Machine, Measurement, Method, Milieu, Material, Market, Money

Figure 1 A generic process model, adapted from Isaksson (2004).

The process model in Figure 1 is also a system model, model in Figure 2 sets out to identify important elements with the different management, operative and support of change. The model extends the division of enablers processes being sub-systems that could be described by and results used in most business excellence models to use of the same model. Especially in larger and more five sets of indicators, which are input, output, outcome, complex organisations there are often several layers of drivers and enablers, Isaksson & Garvare (2003a). The processes found. organisation of indicators into process related groups is As illustrated in Figure 2 the generic process model believed to greatly simplify the assessment of can be used to describe specific organisational systems performance. and their interested parties. The organisational system

Competitive pressure Competition External Enablers (Island location, local attitudes to studies, local (other community interest, organisational base, national application procedures etc.) universities; opting for Organisational mission Stakeholder working life) (to educate, to carry out research and to support to society) pressure

Management Managing performance Processes (Internal Drivers)

External Operative Processes Drivers Providing basic education, researching and supporting society (Market) Stakeholder satisfaction Marketing HGO as an interesting choice (students, society, research Input Output society, employees) (Material) Persuading potential students to apply (Outcome)

Converting applications into enrolled students

Support Providing IT-support Maintaining premises Processes Servicing personnel Improving performance Catering etc. Management Measurement Resources Man Method (Internal Enablers) Machine Milieu Material Money Market

Figure 2 Application of the generic process model on a Swedish university, based on a model adapted from Isaksson & Garvare (2003b).

From a performance assessment point of view, it version of the Fishbone diagram as 7 common causes, could be argued that the first indicators to focus on Bergman & Klefsjö (2003). The two Ms of Market and should be the main parameters for output and outcome. Money have been added. The level of process maturity of When these indicate important improvement potentials the management also form part of the enablers, as do the the reasons could most likely be detected by a careful operational and support processes. Generally the process examination of the external or internal enablers. In Figure maturity level could be seen as part of the Method 2 these are described as organisational resources, resource. categorised by the use of the 9Ms, which originate from a

35 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Suggested Framework for Minimised Business System for 3E indicators Excellence Models The identified stakeholder needs should be converted In our view, the purpose of business excellence into indicators such as customer, shareholder or employee models should be to assess and improve organisational satisfaction or into environmental impacts. The outcome sustainability. The first step in any improvement process parameters, which are stakeholder focused, could then be should be to analyse actual performance. However, translated to organisational Key Performance Indicators before measuring anything it is important to know what for the Output. Provided the improvement potential is to measure. It could be argued that stakeholder needs significant, a review of the enabler performance is should be guiding the choice of indicators, with justified. See Table 3 for a list of criteria for the 9Ms. preference given to the most important needs. One way to Basic questions for the management are if there are objectively derive such needs could be to use the QFD- policies, objectives, responsibilities and measurements based BEST deployment process described by Edgeman for the areas with identified potential. Input measures are & Hensler (2004). In this process principles of SD are of interest when the M of Material contributes to a correlated and prioritized according to Maslow’s substantial part of critical Output indicators. Examples in hierarchy of needs. This is in sharp contrast to the GRI the different dimensions are expensive goods and approach that only looks at how things are done, not on services, materials and input articles with a high what is done. The rating of products based on how environmental load and goods with a questionable social necessary they are, should be an important area for future impact. Buying products produced by child labour would, research. for example, way heavily on the ethical performance of Reporting should be customised and should start with the organisation. The external enablers are such as a listing of stakeholders and their needs. Stakeholders country factors for infrastructure and business climate, could include customers, suppliers, shareholders, employ- which can act as barriers for pressure coming from ees, current and future societies and the nature in general. competition and stakeholders. The external enablers can The process outcome, or impact of the output, is help to explain why both external and internal drivers for stakeholder satisfaction. In the case of Nature, outcome change might be weak. External enablers and drivers are could be measured by indicators such as loss of species or not further discussed in this paper. degraded biotype. The system extension limits for the The five types of measurements could be divided, study should be defined based on some agreed and based on the concept of the Triple Bottom Line, into the measurable organisational impact. In order to define an dimensions of Economy, Environment and Ethics improvement potential the current performance should be (Isaksson & Garvare, 2003a). The reason for using Ethics compared with a system benchmark, relating both to what instead of social responsibility or social equity is to focus other comparable leading organisations are doing and to on the inherent responsibility that every individual, what the actual system performance limits are. The organisation and nation has to contribute to sustainability. detected potential should then guide the focus of the Edgeman & Hensler (2004) take up this by discussing work, into any of the sustainability dimensions or into all kyosei – living and working for the common good. Social of them. A heavy industry might have to start with a responsibility could be seen as giving alms to the poor, limited three-dimensional report where as a well where as we believe that ethics would better describe an performing service organisation could for example start obligation to work with the root causes of poverty and with focus on the social aspects. injustice.

Table 1 A reduced set of generic indicators for process output with additional quality related measurements, adapted from Isaksson (2004). TBL Indicator type Description Comments Dimension Economic Process Cost Full process cost All costs for the product at the organisational interface Process Capacity Relative figure relating Such as Run Factor indication the time the to system capacity process is used and Capacity Efficiency indicating the unit per hour production when process is running Goods Quality Technical quality or Often related to the product specifications what the product can do Service Quality Functional quality or Typical examples are delivery time how the delivery is done Product Sales Value Total value of sales This indicator corresponds to the indicator Specific Product Value of Net sales value in the GRI guidelines. The Specific Product Value is the value for the customer related to the price Environmental Environment (emissions) Emissions to air, water These include the impacts caused by input and land material Social Safety & Health Number of incidents and Mainly related to the stakeholder accidents employees

36 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Ethical performance How the organisation One important aspect is how the demonstrates ethical organisation works for improving its behaviour performance in 3E terms. Another important aspect is how the organisation minimises negative direct and indirect effects from operations.

In Table 1 an example of a limited three dimensional Magnusson & al. (2003). The results could be report is presented. Quality indicators have been summarised as has been done in Table 2. introduced, enabling the calculation of the Cost of Poor Quality (CPQ). According to Sörquist (1998), CPQ is Table 2 Summary of the performance potential – a conceptual defined as “those costs which would disappear if the example. company’s products and processes were perfect”. Using Dimension Potential of Rate of Comment this definition, benchmarks for processes and products performance improvement could be set within the system studied, as the level when improvement everything functions perfectly. Economy Large Low Underperfor mance and ethically questionable MODEL STRUCTURE (secondary potential) Environment Low Low OK The model framework proposed includes assessment Ethics Medium High OK. of organisations in terms of both process performance Additional and resource utilisation. By designing the models based resources on stakeholder needs it is believed that time lost in non could value adding time of the one-size-fits-all approach can be primarily be reduced. invested in other dimensions Performance assessment The performance assessment for output is based on Resource assessment the organisational process model described in Figure 2. The resource assessment is based on the 7Ms The performance assessment has been divided into seven described in Figure 2 with the addition of the M:s of steps: Market and Money, see Table 3. The M of Market

represents the resource of customer base, contacts to 1. Process charting of the organisation using a given, interested parties and position compared to competition. high level process model The M of Money describes the financial resources and 2. Examination of needs in a TBL context - assets that form an important enabler for improvement Identification of stakeholders, stakeholder needs and work. Each criterion should be reviewed based on priorities as well as competition. approach, deployment, results and evaluation & improve- 3. Identification of primary Key Performance Indicators ment. These are all assessed on a scale from 0 to a 100. – the outcome in terms of stakeholder satisfaction – This is similar to the evaluation process of the and assigning them to objectively based priorities. Springboard (Hellsten, 1997). 4. Prioritisation through the translation of Outcome KPI into organisational Output KPI, using Table 2 for Table 3 Proposed resource performance assessment criteria guidance. based on 9M:s. For each M the importance and the rating 5. Defining benchmarks for the Output KPI:s should be assessed. 6. Estimating performance potential as the difference between current performance and benchmarks M:s Criteria 7. Estimating rate of performance improvement by Management Policies, goals and strategies for improving measuring Output KPIs over a longer period of time. TBL performance Method Structure for carrying out strategies and achieving goals. The level of the organisational sustainability Management systems for TBL. Process improvement potential and the rate of improvement maturity. indicate the urgency of change. A significant potential in Man How is relevant competency for the different combination with a low rate of improvement is a sign of work tasks assured weak drivers for change. Furthermore, it signals Measurement How is TBL performance measured and underperformance in the ethical dimension. A benchmark results communicated rate of improvement could be such as realising 50 percent Machine Describe how the equipment including all of the estimated improvement potential yearly. This is premises is made to perform to the required comparable with the Six Sigma benchmark improvement standards of reducing the DPMO-level with 50 percent per year, Material Describe how the material used is monitored and controlled Milieu Describe how the working environment

37 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

supports the personnel and the equipment Product Value Income for the university for the different used courses compared to direct costs; Market Describe market information such as Educational market value for the student (and customer base, market perspectives, society) and the Specific Product Value as the competition and important stakeholder needs educational market value compared to study Money Describe availability of funds for operations efforts and costs. and investments The case description in this paper has a narrow focus The organisation should set priorities and weights for on the two stakeholder categories of university and the different M:s based on the actual situation. Out of the students. In Table 4 outcome requirements have been different M:s, Management is normally the most translated into proposed output Key Performance important aspect to consider and should therefore Indicators for the economic dimension. probably be in early focus. The purpose of using this The survey showed that it was not possible at the model for assessing organisational sustainability should university level to find sufficient aggregated data to be internal comparison over time, not an absolute establish the proposed KPI and only a qualitative comparison with other organisations. Therefore the assessment is carried out. The process cost is highly choice of priorities should be made by the organisation variable between courses and no benchmark or target has itself. Focusing on what is relevant reduces the work been established. As a total the budget goal of a zero load. If some of the M:s are considered irrelevant after an result has not been achieved, which shows that the initial review they should be omitted. average cost per study point is too high. The important capacity parameter is the capacity utilisation of available teachers and classrooms. With a large number of courses with few students the capacity INDICATIVE APPLICATION OF THE utilisation is in the range of merely 30 to 60 percent. The MODEL academic quality varies and is not monitored on a permanent basis. The service quality is monitored but the Case study of a Swedish university surveys are not compiled and results are not presented, which leaves the question of the performance level open. The Business Excellence model for Organisational The product value of the different courses for the Sustainability presented in this paper has been partly university varies from important losses to good profit but tested on a Swedish university. The process chart the details are hidden in the budget figures. The prepared is presented in Figure 2 where also principal educational value has occasionally been put into stakeholders are identified. In the initial evaluation direct questions since a large part of the courses and programs environmental and ethical impacts were judged to be of do not directly lead to any jobs but mainly contribute to less interest than the economic performance. Hence, the student’s personal academic level. focus was placed on the economic performance. Main The market is such that the students choose between stakeholders include: different educational programs and the state pays. This • the state – that pays for all the education means that if the university perceives that there is a • the students – that invest their time and money in their market for a particular type of education, such as multi education media production, show business or game programming, • the employers – that will hopefully hire the students then a program can be launched with good economic in the future results. With possibly a low educational value for society • the university – that must survive in a market and the student, this could indicate dubious ethical orientated environment. behaviour based on short-sighted economic interests. The qualitative performance assessment indicates an Table 4 Proposed Key Performance Indicators (KPI) for the important improvement potential and also gives a clue to educational process output at the studied Swedish university. the performance of some of the enablers. The enablers are briefly reviewed in Table 5. Economic Description and comments dimension (KPI)

Process cost Total cost for a study equivalent (in Sweden a full student year is 40 points and a point could be used as the equivalent and denominator) Process capacity Number of courses and programs and total student study points examined compared to the system capacity or to the original course/program offer Goods quality Academic quality and resource offering of the course within the given topic (product) Service quality How the education is delivered (pedagogy. facilities, study hours, work load etc.). This is measured using course evaluations.

38 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Table 5 Indicative preliminary resource performance assessment with the purpose of presenting an example. M:s Criteria Manage- A recent vision has been written and some policies relating to the 3Es exist but they have not been ment translated to understandable objectives and strategies. At the level of the operational performance very little of the vision can be seen. Importance: High. Rating: 20% Method There is a basic management structure with focus on the economic control. There are no management systems for quality or environment. There is no clear structure for breaking down objectives to the operational level. The process maturity is very low. Importance: High. Rating: 10%. Man For the educational part the competency is relevant and recruitment is closely monitored with focus on academic merits and pedagogical skills. There is little focus on organisational and managerial skills. Importance: Medium. Rating: 40%. Measure- The 3E performance is not properly monitored and presents a major problem. ment Importance: High. Rating <10%. Machine The premises are new and equipment adequate. Importance: Low. Rating 80%. Material The Material in this context consists of the incoming level of new students. Entry levels have been lowered to increase number of students, which increases work load and reduces throughput. Importance: Medium. Rating: 40%. Milieu The physical work environment is good. The social work environment is hampered by stress and organisational problems. Importance: Medium: Rating: 30%. Market The university market in Sweden is highly competitive with an offering exceeding the demand. The university profile has been discussed but not decided, which hampers market focus. Importance: High. Rating: 30%. Money Funds available for development and improvement are limited. Current economic problems and previous deficits have further reduced resources. Importance: High. Rating: 20%.

that mainly are for the self actualisation of students, but Case study analysis that do not contribute to employability, could be seen as The results show that there is considerable potential non sustainable. of improvement, as indicated both by the output performance and the level of enablers. The rate of improvement has not been assessed but an educated guess would probably reveal that due to lack of Money and CONCLUSIONS AND DISCUSSION clear objectives the rate is not up to benchmark performance. A principal conclusion is that the proposed framework Based on the review the most important Ms to focus seems to be relevant for the studied organisation. The on should be: Measurement (<10%), Management (20%), prioritisation and selection steps of the tested model, Money (20%), Method (20%), and Market (30%). An aimed at minimising organisational work efforts, also interpretation is that measurements are needed for appear to be working as intended. However, a number of management to be able to appreciate opportunities and issues still remain for future research. One of these is the assess root causes. The result should be investments in translation of stakeholders needs into prioritised improved organisational methods that would also help to performance indicators for the outcome and then further establish the targeted market. into organisational output indicators. This is a work that The performance level found is believed to be could probably be done using a QFD-based approach as common for many organisations that have not worked proposed by Edgeman & Hensler (2004). The suggested systematically with performance improvement. The generic Economic indicators and the process of setting initial analysis of the case study organisation indicated system related benchmarks are other important areas to that in a 3E context focus should, to start with, be on the look at. This is where the magnitude of improvement Economic performance, but also that this should not be at potential is defined, which guides the degree of effort the expense of Ethical performance. It is important to used to realise the potential. The enablers based on the visualise the indirect sustainability effects, even in the 9Ms should be further developed to cover all relevant cases where these are hard to assess. Note, for example, resources that are required for optimal process the negligible direct Environmental effects of educational performance. activities. Based on this it could be put into question if In quality terms there is a saying which is “to do the work with ISO 14001 or applying the full GRI indicator right thing in the right way”. The first part is about structure would make any significant contribution. having the right products that the customer asks for and However, there could be important indirect the second part is about performing all processes to environmental and other sustainability effects in deliver those products perfectly. When products and education. With a sustainability profile the university processes are perfect the Cost of Poor Quality is zero. could integrate the 3E dimensions in all education to a Translated to a sustainability context the right products larger or lesser degree and thereby have an indirect effect should be those that are based on balanced stakeholder on sustainability through the future work of the current needs. Using scarce resources for the production of students. The content of courses could be reviewed based products for self actualisation is therefore not sustainable. on reference to the Maslow hierarchy of needs. Courses Doing things right would be providing the right things to

39 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 the lowest possible Triple Bottom Line cost. This cost /17/ Isaksson, R. & Garvare, R. (2003a). “Measuring would be the sum of the product price and the sustainable development using process models”, Managerial environmental and social effects. These are complex Auditing Journal, 18 (8), 649-656. issues both politically and technically, but should /18/ Isaksson, R & Garvare, R. (2003b). “Drivers, enablers and barriers towards sustainable organisational performance”, definitely merit future research. Proceedings of the 6th International Conference on Quality The work presented herein is still in an early Management and Organisational Development – focusing on conceptual stage. Our purpose has been to present some Sustainable Development, October 1-3, Paris, France. ideas for further discussions in an area that we believe /19/ Kaplan R. S. (1998). Innovation Action Research: could be of interest to many. Creating New Management Theory and Practice. Journal of Management Accounting Research, 10(March), 89-118. REFERENCES /20/ Karapetrovic, S. (2002). “Strategies for the integration of management systems and standards”, The TQM Magazine, /1/ Adams, C.A. (2004). “The ethical, social and 14 (1), 61-67. environmental reporting-performance portrayal gap”, /21/ Karapetrovic, S. & Jonker, J. (2003). “Integration of Accounting, Auditing & Accountability Journal, 17 (5), 731- standardized management systems: searching for a recipe and 757. ingredients”, Total Quality Management, 14 (4), 451-459. /2/ Aaronson, S. A. (2003). Corporate Responsibility in the /22/ Magnusson, K.& Kroslid, D. & Bergman, B. (2003). Global Village - The British Role Model and the American Six Sigma - The Pragmatic Approach, Second Edition, Laggard, Business and Society Review, 108 (3), 309-338. Studentlitteratur, Lund. /3/ Bergman, B. & Klefsjö, B. (2003). Quality from /23/ Sörqvist, L (1998). Poor Quality Costing, Doctoral Customer Needs to Customer Satisfaction, Second Edition, Thesis No 23, Royal Institute of Technology, Stockholm, Studentlitteratur, Lund. Sweden. /4/ Bryane, M. (2003). Corporate Social Responsibility in /24/ Töpfer, K. (2000). “The triple bottom line economic, International Development: An Overview and Critique. social, natural capital”, UN Chronicle, 36 (2), 39-41. Corporate Social Responsibility and Environmental /25/ Wilkinson, G. & Dale, B.G. (1999). “Integrated Management, 10, 115-128. management systems: an examination of the concept and /5/ CEN. (2000). Quality Management Systems – theory. The TQM Magazine, 11 (2), 95-104. Requirements (ISO 9001:2000). SIS Förlag AB, Stockholm. /26/ Zairi, M. & Peters, J.(2002).”The impact of social /6/ Edgeman, R.L. & Hensler, D.A. (2001). “The AΩ responsibility on business performance”, Managerial Auditing chronicle: Earth@omega or Sustainability@alpha?”, The TQM Journal, 17 (4), 174-178. Magazine, 13 (2), 83-90. /7/ Edgeman, R.L. & Hensler, D.A. (2004). QFD and the BEST Paradigm: Deploying Sustainable Solutions, Proceedings from World Review of Science, Technology and Sustainable Development Second International Conference, Edinburgh, Nov 8-10. /8/ Elkington, J. (1999). Cannibals with Forks: The Triple Bottom Line of 21st Century Business, new ed., Capstone, Oxford. /9/ Eriksson, H. & Garvare, R. (2005). “Organisational performance improvement through quality award process participation”, accepted for publication in the International Journal of Quality and Reliability Management. /10/ ESI (2002). Environmental Sustainability Index 2002, Center for International Earth Science Information Network, Columbia University, NY, www.ciesin.columbia.edu/indicators/ESI. /11/ Garvare, R. & Isaksson, R. (2001). “Sustainable Development – Extending the Scope of Business Excellence Models”, Measuring Business Excellence, 5 (3), 11-15. /12/ GRI. (2002). Global Reporting Initiative – Sustainability Reporting Guidelines. Global Reporting Initiative, Boston. http//:www.globalreporting.org. /13/ Hellsten, U (1997). The Springboard. A TQM-Based Tool for Self-Assessment, Licentiate thesis 1997:47, Department of Business Administration and Social Sciences, Division of Quality Technology & Statistics, Luleå University of Technology, Luleå. /14/ Hellsten, U. & Klefsjö, B. (2000). “TQM as a management system consisting of values, techniques and tools”, The TQM Magazine, 12 (4), 238-244. /15/ Hendricks, K.B. & Singhal, V.R. (1997). “Does implementing an effective TQM program actually improve operating performance?”, Management Science, 43 (9), 1258- 1274. /16/ Isaksson, R (2004). Total Quality Management for Sustainable Development - Focus on Processes, Doctoral thesis, 2004:18, Division of Quality and Environmental Management, Luleå University of Technology, Luleå, Sweden.

40 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

QUALITY MANAGEMENT MATURITY AND MANAGEMENT ATTITUDE TO QUALITY

1) Mr Milan Ivanović, 2) Prof. Dr. Vidosav D. Majstorović, 1)LR, Belgrade 2)Mechanical Engineering Faculty, Belgrade

Summary: Establishing maturity of processes' management (quality is just one aspect), is connected with the model of evolutionary development, towards to the condition of maturity-capability of continual improvement (TQM). Result of assessment is establishing position of management in relation to the defined form of maturity (accepted model of TQM). Experience shows that the relationship of management towards the quality management is very important, since that is the fact which determines further development of this model, and basically, the organization too. Starting from the mentioned, it is very important for every organization to establish level of process’s maturity, in order to continue to the development and implementation of the continual improvements model, especially from the place and the role of the management. In this paper we shall presenting the model for assessment of the process’s maturity that encompasses the relationship of the management towards the quality management. Key words: Management attitude, Management maturity, Quality.

1. INTRODUCTION Table 1. Criteria for the assessment of maturity/capability of Establishing of the level of quality management’s the process maturity is basically founded on establishing the maturity Gra Attributes of process from the aspect of of the processes’ management (management processes de quality output of processes and operating processes). If we talk only about quality (in (product of the process) narrower and broader sense), than we only observe aspect 0 Process is not being applied of processes’ management that refers to the quality. 2 Process is not formalized and depends on the Which processes will be relevant for the assessment, engagement of the individuals, process is not depends from the projected approach. Possible ways from effective the aspect of development and implementation of system 4 Process is disciplined (schedule of activities approaches are: (i) management through inspection, (ii) is obeyed ), repeatable, approach reactive, management through quality control, (iii) management (reacts only when problems appear) through quality assurance, and (iv) management through 6 Process defined (formalized ,documented), total quality. Going towards TQM, number of processes process effective, that are encompassed, is increasing, in the direction from approach proactive inspection processes to all processes in organization. In 8 Process controllable, results predictable, projecting and implementation of the processes’ control of resources, process efficient management maturity model (organization can also be 10 Process is subject of continual reviewing, shown as a process), it is most important to identify improving and transforming definition of maturity. In the most known model for the assessment of maturity - CMM (Capability Maturity Remark: Effective process should mean process Model), used in software industry, following definition that is capable to achieve its purpose. Efficient process was developed : Process maturity is the extent to which a achieves its purpose with the optimal use of resources. specific process is explicitly defined, measured, Results of research have confirmed basic hypothesis, controlled and effective. It is specially mentioned that this but “illogical of results” problem was spotted and it assumption was foundation for the definition of processes shows the need for additional supplement in the in standard ISO 9000:2000. With development of definition of the management’s maturity in certain sense. organizational processes there was need to include continual improvement in definition, so the new definition was: the extent to which a process is explicitly defined, managed, measured and continuously improved. 2. SPOTTED PROBLEMS DURING Starting from this definition of maturity of processes’ management, a model showed in table no.1 was RESEARCH OF THE MANAGEMENT’S developed and practically applied for research of the level LEVEL of quality management in the countries of the West Balkans /7,10/. Results of researches show that processes management in “average certified organization” /7/ is on the level that is corresponding to the condition when processes are: defined, documented and effective.

41 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

However, processes that refer to the introduction of Similar problems are already spotted, because in TQM, benchmarking and management of human organizations with high level of organizing, the resources are on significantly lower level and mainly management’s relationship becomes dominant, especially represent processes that depend on the individual’s when management is based on connecting of strategically engagement or are being carried out sporadically. Starting and operating management, i.e. when the strategy is from the fact that mentioned processes are key element connected with every-day activities. Researches /4/ show for transfer to the higher level of organizational the need for expanding of definition of processes’ capability, i.e. condition of continual improvement, so the management maturity, and to include the organization’s knowledge of good practice’s principle in quality attitude in it. New definition of the maturity of the management isn’t on that kind of level to make further processes’ management is: The degree to which processes development possible. and activities are executed following “good practice” “Illogical of results” problem of the situation, i.e. principles are defined, managed and repeatable grade, is in the fact that lower levels of the quality (institutionalized). From this definition derives necessity management are possible to achieve with old-fashion to identify principles of good practice, and after that, technology and obsolete methods of work. In practical define processes, control them and improve. Second thing sense it is possible to document present practice that is which is very important here is explanation of the term not in accordance with principles of good practice, and to “repeatable-institutionalized”. For the purposes of this regularly apply and carry out corrective actions by means paperwork, we can use following explanation /4/: of solving problems and maintaining of present condition, “Institutionalization entails building an infrastructure but in that relationship of the management and the quality and a corporate culture that supports the methods management there are no pre-conditions for further practices and procedures of the business so that they development. endure after those who originally defined them have In this way, received grade provides more suitable gone”. Term of corporative culture is closely connected picture that it is in reality. From here comes the with leadership, i.e. making of climate in organization conclusion that scale for assessment of the management’s that is necessary for achieving of predicted vision, level should be corrected in order to encompass the mission, i.e. implementing of defined strategy. relationship of the management and the principles of In order to measure till which level have the good practice of quality management. principles of good practice been implemented, it is necessary first to define principles themselves. Regarding principle TQM there is no theoretical consensus /5/ and in author’s opinion, it is the matter of organization to accept 3. ATTITUDE OF TOP MANAGEMENT TO and ponderous (determines significance of an individual QUALITY MANAGEMENT principle) principles in accordance with set up vision and strategy. The most often used and quoted principles are Before analysis and involving of the top the ones under which the European model of business management’s relationship towards the quality in the excellence was developed (EFQM), and they are: (i) scale for assessment of the maturity, it is necessary to results orientation, (ii) customer focus, (iii) leadership explain the term “attitude”. According /9/ it means: and constancy of purpose, (iv) management by processes Attitude is an enduring disposition to consistently and facts, (v) people development & involvement, (vi) respond in a given manner to various aspects of the continuous learning, and (vii) public responsibility. world, composed of affective, cognitive and behavioural The table 2 shows how the mentioned principles components. It is also necessary to explain the evolve during applying in the practice, i.e. how the components of “the attitude”. Affective component is maturity is achieved in applying of certain principles. component of attitude that reflects an individual’s general feelings or emotions toward objects. Cognitive component is component of attitude that represents one’s awareness of knowledge about an object. Behavioural component is component of attitude that reflects buying intentions and behavioural expectations reflects a predisposition to action. So, we can conclude that relationship of the management towards the quality management depends on emotions, knowledge and predisposition for action, or simply said, relationship depends on belief that through methods and techniques of the quality management, organization’s performances could be improved. If we go back to the research’s results, it could be said that on the lower levels of the quality management more important thing is relationship of the management towards the principles of good practice in the quality management, than formalizing of present condition that has no assumption of further development.

42 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Table 2. Example of the maturity model applied in EFQM model of business excellence No FUNDAME- BEGINNING DURING MATURITY NTAL CONCEPT Focusing on the All actions’ leaders Needs of actions’ leaders Transparent mechanism 1. results are identified assessed in structured way needed for balanced expectations Customer focus Customer Objectives connected with Influence for achieving of satisfaction is needs and expectations of customers’ needs and their evaluated customers. Reviewed loyalty. loyalty. Understood, 2. measured and controlled. Leadership and Vision and mission Policy, people and processes Mutual values and ethical consistency of the are defined are aligned to the vision. rules exist at all levels 3. purpose Model of leadership exists Management Processes Use of information for review Capability of the process through the necessary for of predicted objectives understood and used for 4. processes and data achieving of moving of improvement of objectives are performances defined Development of Employees accept Employees are innovative Employees are capable to employees and ownership and and creative in achieving of gather and share knowledge 5. their involvement responsibility for the objectives and experience solving of the problems Continual Identifying of Continual improvement is Successful innovations and learning of opportunities for accepted as an objective of improvements are spreader 6. innovation and improvement and every individual and integrated improvements action 7. Development of Process for Improvement of suppliers Organization and key partnership selection and and achievements are partners are mutually relations control to the recognized and also key independent. Plans and suppliers exists external partners policies are developed on the basis of mutual values. Public Legal and There is active involvement Social expectations are 8. responsibility. regulation in the society. measured and achieved. requirements are identified and achieved.

Here is very important to mention that these are this, there comes need for careful selection and principles that are adequate for organizations that are ponderising (establishing of significance) principles of aimed for the excellence in industrially developed good practice for an individual organization. countries. Theoretically, it is known that employees in organizations with low salaries accept autocratic models of management better than democratic, while in developed organizations with high salaries more 4. DEVELOPMENT OF THE NEW SCALE appropriate method is democratic method of FOR EVALUATING OF THE PROCESSES management. Explanation should be found in MANAGEMENT’S LEVEL Masloviev’s scale of human needs, i.e. poor people search for social security, protection in their leader, while Starting from the first scale and mentioned references, socially secured people aim for their own development a new scale was developed, shown in table no. 3, that is and emancipation. Also, theoretically known fact is that based on definition that respects the relationship of the depending on the phase of life cycle of the product organization towards the applying of the principles of (company), there are various adequate styles of good practice, so the level no. 5 in this scale means that management. In starting and mature phase of the the principles of good practice are basically part of development, autocratic styles are adequate, while in the corporative culture. phase of growth the most appropriate is the democratic style of management.

It is a special question of compatibility of the national culture and culture that is to be imposed in the organization through the vision and strategy. From all of

43 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

LEVEL OF ATTRIBUTES OF THE PROCESS 5. CONCLUSION PROCESS FROM THE ASPECT OF QUALITY MANAGEMENT 1. Process is not defined, dependent from Expanding of the criteria for the assessment of the the individual, not effective, measuring management’s level, by introducing of the relationship of of quality at the end of the process. the top management towards the quality by means of Employees are not aware of the use of capability to accept principles of good practice, principles of good practice. Studying management’s maturity is connected to certain theoretical reactive, on errors. models in the shape of principles. In this way, model Improvements are aimed on standardization of the product and provides the maturity’s grade in relation to the applying technology - SDCA cycle. of the good practice’s principles, and not till what extent 2. Mostly functional processes, is the current condition defined, controlled and effective. disciplined. Measurements at the end Concrete consequence on results received in research and in individual phases. Individuals would be seen in following, that the real level of the are aware of the use of applying the management in our organizations is observed from the principles of good practice but aspect of the good practice’s principle, in much lower processes are not repeatable. Studying level, than the one the research has shown. However, the is planned. Improvements are aimed only thing that is clear now is why there is no continual towards the standardization of the process’s results - SDCA cycle. improvement in our organizations. We will discuss about 3. Process formalized and effective. details of that kind of analysis in some of the next paper. Measuring in all process phases and review if the formalized process is being applied. Management aware of the use of applying of principles of REFERENCES good practice. Learning aimed toward competency. /1/. Tenner, A., Process Redesign, Addison – Wesle, Longman, Improvements aimed towards the London, 1997. standardization of process itself - /2/. Leo, T., Shiba, S., Integrated Management Systems, John SDCA cycle. Wiley & Sons Inc., New York, 1999. Process managed from the aspect of use 4. /3/. Majstorović, V., One model for assessment business of the resources, optimized and excellence of manufacturing systems, Proceedings of Second efficient. International Working Conference ''TQM & AIA'', pp. 21-25, Measuring of resources’ usage i.e. Kragujevac, 2003, process’s efficiencies. Everyone is /4/. Moultrie, F., Gregory M., The use of maturity models/grids aware benefit of the use good practices as a tool in assessing product development capability, IEEE, principles. Studying is aimed toward Cambridge, 2002 strategically objectives, quality aspect, /5/. Majstorović, V., Model of the total quality management, product’s improvement and process Bussines Politic, Beograd, 2000. from the aspect of customer satisfaction /6/. Moore, Ch., The learning management maturity model, - PDCA cycle. ATHINQ White Paper, 2002. 5. Process loses its significance because /7/. Ivanovic, M., Majstorovic V., Research of the quality whole organization is integrated around management’s level in the certified organizations of our achieving of the vision. country, Balkans Conference on Quality “THE BALKANS AS Performances of the organization are A REGION OF QUALITY”, Belgrade, 2005. being measured, quality aspect. /8/. Ivanovic, M., Majstorovic, V., Analysis of the model for the Awareness of the importance of good assessment of the quality management’s maturity, Scientific- practice’s principles is completely expert Journal ''Total quality management'' No2 Vol.32, rooted in the organization. Beograd, 2004. Fundamentally important, corporative /9/. William G. Zikmund, Business research methods, The culture for the success. Dryden press, Philadelphia, 2000. Learning is aimed towards the balanced /10/. Ivanovic, M., Majstorovic, V., Developed Model for the strategy (stakeholders). Assessment of Quality Management Level in Manufacturing Improvement of the process’s System, International Conference on Quality, Tokyo, 2005. management - PDCA cycle. /11/. Ivanović, M., Development of the model for the assessment of the quality management’s level in production systems, Table 3. New scale for the assessment of maturity/capability of (Doctorate thesis in working progress), Beograd, 2005. the process In the showed model for the assessment of maturity of the processes’ management, it is very important to notice, that on the fifth level of maturity, an integration of the management in the organization is being achieved through making and implementation of corporative culture for the aim of achieving of the predicted vision. Since the whole organization is integrated (IMS), process management (process approach) loses its meaning.

44 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

DESIGN OF A SYSTEM FOR INTELLIGENT DATA-POINT PRE-PROCESSING IN REVERSE ENGINEERING Keynote paper

I. Budak a), M. Sokovic b), J. Hodolic a) a) Faculty of Technical Sciences, University of Novi Sad, b) Faculty of Mechanical Engineering, University of Ljubljana,Slovenia

Summary: Reverse Engineering, as a solution for complex shapes’ design, has been developing very fast during the last decade. However, even with the fast development over the last few years, Reverse Engineering still has certain drawbacks. One of the main problems is related to the low quality of 3D-digitization results (noise, measurement errors, redundant data, incomplete data, etc.), which implicates less accurate surface reconstruction. Solving, or at least reclamation, of this problem is possible by preparing the 3D-digitization results through data filtering, data smoothing and data reduction. These steps are, among some others, normally implied in pre-processing phase of the Reverse Engineering process. On the other side, we are witnesses of artificial intelligence development and of its implementation, in different shapes, wherever automation of decision making process on the basis of previous experience is needed. In that sense, the subject of this paper is a presentation of possible improvements through the implementation of artificial intelligence elements on the developed conventional system for pre-processing in RE. Keywords: Reverse Engineering, data pre-processing, noise filtering, data reduction

1. INTRODUCTION Neither RE is excluded from this. On the contrary, we can say that the process of RE is very convenient for Without any doubt, Reverse Engineering (RE) is application of different artificial intelligence methods. becoming a significant tool within the process of CAD Firstly, this is because of the great need for decision product modelling, especially when concerning complex making and acting on the basis of previous experiences, surfaces, which are often very difficult or almost whether it is in surface reconstruction, geometric feature impossible to model by using tools offered in modern recognition, or deciding adequate methods and CAD systems. Thus, those shapes can be made by parameters application, and the like [5,6,7]. Examples of ”artistic modelling” (in clay, gypsum, polymer materials, artificial intelligence application in the field of RE can be wood, etc.), and then finished physical models can be found in [6], where Q. Peng and M. Loftus present an RE transformed, by RE, into CAD models. Furthermore, RE system with image processing based on neural networks. application in the redesigning process of either one's own Neural networks are used for shape recovery on the basis or competition's product can significantly accelerate and of illumination parameters. It can be said that pre- improve the process itself [1]. In the case of necessity for processing is done through a process of training neural producing the copies of parts and products without network, when it “learns” how to treat the gained adequate technical documents, RE is almost certainly information. without any alternatives [2, 3]. Also, Jun et al. present in [7] an RE system based on RE, until about ten years ago, presented a notion not neural networks for geometric feature recognition. In this so often found in literature, and shyly used in practice. system, surface is reconstructed by extraction of data- Today, the reverse engineering technique is commonly point groups that match certain geometric feature (hole, applied in many fields: production engineering, software groove, round, etc.). Matching is done by neural network. engineering, film industry, entertainment industry, The main subject of this paper is to perceive chemical engineering, electro-technical industry, and possibilities for application of artificial intelligence in recently, there are first examples of RE application in order to improve one developed system for data-point industry of MEMS (Micro Electro-Mechanical Systems) pre-processing in RE. In that sense in extension of this [4]. paper, after some shorter review of pre-processing in RE, In the same time, methods of artificial intelligence, developed system for conventional pre-processing with such as artificial neural networks, genetic algorithms, emphasized weaknesses will be presented. Finally, at the expert systems, etc., are more and more applied for end, a model of proposed improved system for intelligent solving difficult problems in technique, as well as in pre-processing will be given. other fields. Reason of their application is mainly connected with needs for the automation of decision making process, especially in cases where it is based on some previous experience. Artificial intelligence application raises efficiency and quality by time saving, as well as through human factor elimination [5].

45 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

2. DATA-POINT PRE-PROCESSING IN algorithm based on scanned curves’ straightness is REVERSE ENGINEERING proposed. K. H. Lee and H. Woo have proposed in [12] a novel The starting point of RE-process is a physical object procedure that integrates techniques of RE and Rapid which can be an existing part or a product, a handmade Prototyping (RP) technology. This procedure implies pre- (sculptured) model or a model gained from some processing based on algorithm for cross-sectional data technique for Rapid Prototyping [3]. This physical object reduction on the basis of curvature calculation between is, through the phases of 3D-digitization, pre-processing neighboring points. and surface reconstruction, transformed into a CAD Carbone et al. have presented in [13] a combination of model suitable for usage in modern CAx technologies [1]. a vision system and coordinate measuring machine However, this process is not at all simple. Even though (CMM), for the RE of free-form surfaces. Point-clouds the technologies included in the RE-process have been are pre-processed by means of suitably developed significantly improved in the last couple of years, there procedures for ordering and data reduction. For the are still certain problems influencing significantly the purpose of reducing large number of point-data, two quality of the process. One of the most important algorithms were presented. Random algorithm allows problems concerns the quality of 3D-digitization results, bends to be deleted depending on the probability known as point-cloud, and it considerably influences the specified by the user. It is very useful when a large quality of the resulting model [8, 9]. number of points have to be eliminated in areas of a A large number of different 3D-digitization systems smooth shape. Second data reduction algorithm is based have been developed, among which the most prominent on tolerance of the change of tangent or curvature. are coordinate measuring machines (CMM), laser K. H. Lee et al. introduced in [14] an RE procedure scanners, pantographs, CCD cameras, computer with strong accent on pre-processing phase. After tomography (CT), etc. [2]. Common to them all is the removing outliers and spikes from the initial point-cloud, presence of noise and measuring errors in the results, reduction algorithms are applied. Proposed algorithms are which most often implies generating inadequate model. based on one- and bi-directional non-uniform grid Also, modern 3D-digitization systems (especially laser methods. based systems) generate a large amount of points, which M. C. Huang and C. C. Tai have offered an RE on one side can considerably slow down, and in extreme system with a pre-processing procedure [15]. The concept situations also prevent data-point processing, while on the of the procedure is to regress a set of data-points into other side too large number of points can as a result have non-parametric equation, in implicit or explicit form; this inadequate surface model. Therefore, before the surface equation also has to satisfy the continuity of the reconstruction process, it is necessary to prepare the curvature. Noise data are eliminated by manual action of point-clouds. This preparation is usually described by the the user, and for data smoothing median filter is used. term pre-processing [4, 8, 9]. Up to applied digitization D. Y. Chang and Y. C. Chang proposed in [8] an RE system, RE methodology or surface reconstruction developing flow, where most attention is focused on pre- methods, pre-processing phase can imply different processing phase. Noise treatment is performed by processes, such as: noise and error filtering, data volumetric filter, and for data reduction height decision smoothing, data reduction, data segmentation, data method, which integrates the space concept and height regression, etc. Among them, noise filtering, data characteristic of data-points, is proposed. smoothing and data reduction are most often present steps As an RE system with intelligent pre-processing in this phase [1, 4, 8, 9]. functions we can mention RE approach for complex During the last decade, a great number of RE systems shapes, based on genetic algorithm for the reduction of has been developed. Nevertheless, just in very few of data-points, presented by G. Percoco and R. Spina in them greater attention has been focused on pre-processing [16]. The algorithm retrieves the best population made of functions [8, 9]. In the following part of this section short a group of scanned points fixed by the user, which overview of several characteristic RE systems, from the represent the shape with minimal loss of precision. pre-processing point of view, has been presented. Thompson et al. have introduced in [10] an interactive RE system – REFAB, which allows a user to interactively define a model composed of mechanical features from a 3. DEVELOPED SYSTEM FOR DATA-POINT set of 3D surface points. The problem of pre-processing PRE-PROCESSING is over-bridged by recognizing standard manufacturing features in the point-cloud. Of course, the application of Some of the leading problems, within RE process, this solution is limited only on parts possessing standard concern the quality of 3D-digitization results that mechanical features. significantly complicates processing and surface G. J. Wang et al. have proposed in [11] an RE system generation. As well, in most of the contemporary CAD- for designing sculptured surfaces based on four-axis laser software systems with built-in modules for surface scanning system, where much attention is focused on the reconstruction, functions for data-point pre-processing pre-processing of 3D-digitization data-points. Impulse- are either very poor or do not exist at all [5, 8, 9]. This like noise is eliminated by the usage of a median filter, research is, regarding the above, focused on the and as for redundant data elimination, an intelligent development of a system for pre-processing of 3D- digitization results. In this section a developed system for

46 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 conventional pre-processing will be presented. This will 3D-DIGITIZATION involve the analysis of weak points that should be RESULT improved. The overall algorithm of the developed system is DATA PREPARATION presented in Figure 1. As basic elements of the suggested system, the following can be singled out: Adaptation of input data • Data-point preparation, • Error point filtration, Pre-filtering by Volume filter • Data-point smoothing, • Data-point reduction and Determination of scanned • Output file-format generation. curves' total number (i =1,...,n)

The system for data pre-processing developed within this research is based on the cross-sectional RE- i = 1 methodology. It enables data-points obtained from 3D- digitization system to be accepted, adequately prepared, SELECTION OF i and adapted for the reconstruction in a CAD-model. SCANNED CURVE Developing this module, prominent methods have been used. For filtering - volume and angle methods; for data smoothing - median and average value methods; and for data reduction - uniform sampling, spatial, tangent NOISE FILTERING and straightness methods. By embedding several methods Angle method Median method for same functions, a possibility for the application on data-points with different characteristics received from different 3D digitization systems is ensured. Application of different methods in filtering measuring errors as well as in data-point smoothing does DATA SMOOTHING not have a greater effect on the final result. Average value Median method On the other side, when we are speaking of data- method points reduction, selection of method is much more important and can have a decisive influence on final outcome of RE process. This can be illustrated by comparing data-reduction results obtained through DATA REDUCTION application of different methods. As input data, points of th the 70 scanned curve from the 3D point-cloud of Uniform Spatial Chordal Straightness digitized engine cover of the Volkswagen’s car model sampling method method method from the ’60s - “Karmann-Ghia”, have been used. Figure 2 shows a 3D-digitization result of contact scanning gained from commercial system Cyclone 2, Renishaw. GENERATION OF i Parallel review of the results obtained with four SCANNED CURVE embedded methods for data-reduction, divided in four areas for easier analysis, and is presented in Figure 3. This figure clearly shows differences in points’ arrangement as well as in the amount of points within OUTPUT FILE-FORMAT GENERATION regions. These differences are systematized in Table 1 IBL format PTS format with a review of reduction parameters with applied values, as well as of obtained results in number of points, for both analyzed cross-sectional curve and entire data. On the bases of the obtained results, one can conclude i = i +1 that different data-reduction methods give different results in regions with different shape of data-points. This can affect the process in two ways. First one is related to no i > n the fact that application of some method on different yes scanned curves, within the same point-cloud, can greatly differ according to their shape. Second effect is even PRE-PROCESSED 3D-DIGITIZATION RESULT more sophisticated and is connected with a complexity of scanned curves in the sense that one method can give different quality in different regions within the same Fig. 1. Algorithmic flow of the developed programme system for scanned curve. data pre-processing

47 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

a)

b)

c)

d)

e)

Fig. 2. 3D-digitization result (7652 points)

Fig. 3. Parallel review of data-reduction results of the 70th scanned curve: (a) Original data-points (109 points); (b) Uniform sampling (55 points); (c) Spatial method (56 points); (d) Chordal method (57 points); (e) Straightness method (77 points)

Table 1: Review of reduction methods’ parameters with the obtained results Result of reduction Data-reduction Reduction Parameter [number of points] method parameter value 70th curve Entire data Original data / / 109 7652 Uniform sampling Sampling rate 2 55 3848 Spatial method Spatial distance 2 [mm] 56 3946 Chordal method Chordal height 0,015 [mm] 58 4294 Straightness method Straightness level 2o 77 5446

4. DESIGNING A SYSTEM FOR main improvements are related to functions for data INTELLIGENT DATA-POINT PRE- smoothing and data reduction. PROCESSING The improvement concerning data-point smoothing is in connection to taking into account the roughness level of the scanned surface. The developed conventional system for data-point pre- The central place in the intelligent pre-processing of the processing has presented satisfactory results during the proposed system is held by the data reduction segment. application. However, the employment of this type of As it can be noticed in the model, there is the notion of system requires large experience on the side of the user; it developing artificial neural network, which would be is also demanding in concern to the time needed for trained to choose specific reduction methods dependant defining the methods separately for each curve. on the data-point characteristics (shapes and resolution). Furthermore, this system does not provide the possibility It is expected that one or more additional methods should to alter methods within one curve. In this sense, the be joined to the ones already built in the conventional automation of the choice of methods would significantly system. Certainly, before anything else, it is necessary to improve the efficiency and process quality following the convey the analysis of method applications with data- decrease in time and elimination of the human factor points of certain properties. errors. Considering the fact that the automation is conditioned by decision making based on certain knowledge and experience, with the need of their development, the application of artificial intelligence 5. CONCLUSION imposes itself as the tool for improving the existing system. Figure 4 shows a model of the proposed system for In order to obtain adequate CAD-model, since surface inteligent data-point pre-processing. As it can be seen, reconstruction on the basis of raw 3D-digitization data often results in unsatisfactory CAD-model, pre- processing data-points before surface reconstruction is

48 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 needed. During the last decade, a great number of RE Firstly, a developed system has been shown, where the systems has been developed, yet just in very few of them emphasis has been placed onto the drawbacks that should greater attention has been focused on pre-processing be perfected, and then a model of an intelligent pre- functions. processing system has been introduced. Within the However, lately a lot of effort has been consumed into proposed improvements, there is also the proposal to the development of a pre-processing system, whether as develop neural networks for the intelligent choice of data an integral part of the 3D digitization system, or as a reduction methods. module within the software for surface reconstruction. The realization of such a system would certainly raise the quality of the final results with the significant 3D-DIGITIZATION decrease in time necessary for pre-processing. RESULT Furthermore, the demand for greater experience of the DATA PREPARATION user would be eliminated.

Adaptation of input data

Pre-filtering by Volume filter ACKNOWLEDGEMENTS

Determination of scanned The authors would like to thank Dr. Jorn Mehnen curves' total number (i =1,...,n ) from University of Dortmund - The Department of Machining Technology for the 3D digitization results i = 1 used in this research as the input data.

SELECTION OF i SCANNED CURVE REFERENCES

NOISE FILTERING /1/ Varady T., Martin R.R. & J. Cox, Reverse Choice of parameter Engineering of Geometric Models - An Introduction, Computer DATA SMOOTHING values based on Aided Design, Vol. 29, No. 4, April 1997, pp. 255–268. roughness level /2/ Sokovic M. and Kopac J., 2003: RE (Reverse Engineering) as Necessary Phase by Rapid Product INTELLIGENT Development, Proceedings of 12th Int. Conference AMME DATA REDUCTION ´2003, Zakopane Poland, December 7-12, pp. 825-830. /3/ Duhovnik J. & Tavcar J., 2002: Reengineering with CURVES’ SHAPE ANALYSIS rapid prototyping, TMCE 2002, Proceedings of the fourth international symposium on tools and methods of competitive engineering, April 22-26, 2002, Wuhan, P.R. China. Wuhan, DATA REDUCTION China: Huazhong University of Science and Technology Press, 2002, pp. 117-130. /4/ Budak I., Hodolic J. & Sokovic M., 2005: Development of a programme system for data-point pre- i = i +1 processing in Reverse Engineering, Journal of Materials Processing Technology, Elsevier (article in press). no /5/ Jockovic M., Ognjanovic Z. & Stankovski S., 1997: i > n Artificial intelligence - Intelligent machines and systems, Krug, yes Belgrade (in Serbian) /6/ Peng Q., Loftus M., 2001: Using image processing SCANNED CURVES ASSEMBLE IN 3D POINT-CLOUD based on neural networks in reverse engineering, International Journal of Machine Tools & Manufacture, Vol. 41, pp. 625- 640. PRE-PROCESSED /7/ Jun Y., Raja V. & Park S., 2001: Geometric Feature 3D-DIGITIZATION RESULT Recognition for Reverse Engineering using Neural Networks, The Int. Journal of Advanced Manufacturing Technology,

Vol.17, Springer-Verlag London Limited, pp. 462-470. Fig. 4. Model of the system for intelligent /8/ Chang D.Y. & Chang Y.M.: A Freeform Surface data-point pre-processing Modelling System Based on Laser Scan Data for Reverse Engineering, The Int. Journal of Advanced Manufacturing The demand for faster and more qualitative RE Technology, Vol.20/1, July 2002, pp. 9-19. process and together with it data pre-processing, in one /9/ Lee K.H., Woo H. & Suk T.: Data Reduction hand, and faster development of artificial intelligent Methods for Reverse Engineering, The Int. Journal of Advanced systems and their employment into almost all spheres of Manufacturing Technology, Vol. 17, No. 10, May 2001, pp. technique in the other hand, led to their application in this 735-743, ISSN 1433-3015. field as well. /10/ Thompson W. B., Owen J. C., de St. Germain H. J., Therefore, this paper has presented the proposal for Stark S. R., Henderson T. C., 1999: Feature-Based Reverse Engineering of Mechanical Parts, IEEE Transactions on improving one developed system for data-point pre- Robotics and Automation, Vol. 15, No. 1. processing by applying artificial intelligence elements.

49 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

/11/ Wang G.-J., Wang C.-C., Chuang S.H.F., 1999: Reverse Engineering of Sculptured Surfaces by Four-Axis Non- Contacting Scanning, The Int. Journal of Advanced Manufacturing Technology, Vol. 15, pp. 800-809. /12/ Lee K.H., Woo H., 1998: Use of Reverse Engineering Method for Rapid Product Development, Computers Industrial Engineering, Vol. 35, No 1-2, pp. 21-24. /13/ Carbone V., Carocci M., Savio E., Sansoni G. and De Chiffre L., 2001: Combination of a Vision System and a Coordinate Measuring Machine for the Reverse Engineering of Freeform Surfaces, The Int. Journal of Advanced Manufacturing Technology, Vol. 17, pp. 263-271. /14/ Lee, K.H., Woo H., Suk T., 2001: Point Data Reduction Using 3D Grids, The Int. Journal of Advanced Manufacturing Technology, Vol. 18, No 3, pp. 201-210, Springer-Verlag London Limited. /15/ Huang M.-C., Tai C.-C., 2000: The Pre-Processing of Data Points for Curve Fitting in Reverse Engineering, The Int. Journal of Advanced Manufacturing Technology, Vol. 16, pp. 635-642. /16/ Percoco G. and Spina R., 2001, A Genetic Algorithm Approach for the Reduction of Point Clouds of Scanned Complex Shaped Parts, Annals of DAAAM for 2001, Vienna, Austria, pp. 355-356.

50 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

„FAST FORWARD” – DESIGNED PROCESS OPTIMISATION

Prof. Dr.-Ing. R. Schmitt; Dipl.-Ing. J. Dören Fraunhofer Institute for Production Technology, Department Metrology and Quality Management, Aachen, Germany

Summary:In modern production, products have to be with good quality, even if the price is low. Therefore every production process has to be optimised concerning speed and quality. Especially for complex processes it is difficult to reach these contradictory targets. The thermal spraying process is an example for this kind of complex processes. Key words: Quality, DoE, Improvement.

APPROACHES TO BE “BEST IN CLASS” method continuously. 32.4% use it rarely and 49% do not use it at all despite knowing the method /1/. One reasons The methods of DoE are different mathematical ways for this result is that the method seems to be very difficult to plan and analyse experiments. In industry DoE is used and complex. Some companies assess the effort to apply to analyse and optimise processes or products with DoE as to high, because it is to complex and to many mathematical support. But by using only DoE, the experiments have to be run. The reason for this is that the optimisation of processes and products will not be very DoE is often used without additional methods like system efficient. In a survey of the Fraunhofer IPT in 2002 4000 analysis or others. companies of the areas automotive, precision and By supporting it with a project management and other mechanical engineering, plant construction and electrical quality management methods, the experimental engineering were asked about the application of QM- complexity can be minimised. Methods and their success for the company. 443 results The first step to make DoE more efficient is to add a were used in the study showing that DoE is known by system analysis to DoE and to structure the procedure of 44,2% of the interviewed companies. But although DoE using DoE Fig. 3 shows this procedure and its steps. is well known, only 18,5% of the companies use the

System analysis Experimental strategy

„ Definition of objective AB „ Select an experimental „ Collection of potentially 1-- design important 2+ - „ Define factor levels - product attributes 3-+ - process parameter 4++ „ Number of runs

Analysis of results Performance of experiments „ Compilation and presentation of statistical variables Performance„ Conduct of experiments experiments -Effect - Interaction „ Record results „ Interpretation of the results

Fig. 3: Procedure of applying DoE. /2/

The first step is a system analysis to identify all factors, factor interactions and mathematical models relevant factors. Depending on the number of factors and aiming to improve/optimize processes/products are the objective of the procedure a design plan is chosen. determined. After conducting the experiments, the results are analysed Most industries do not use this approach. They always with help of statistical methods. This way the significant try to start by designing the experiments without knowing

51 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 anything about their relevant factors, responses and formulated and implemented (Do). The result must be targets of the experiments. As a result too many evaluated, compared with the previously defined target experiments are made or wrong factors / wrong factor (Check) and if required, further improvements must be levels are used. So the analyse of the experiments will be made (Act). faulty. DoE is one important method of the SixSigma For DoE it is important to take some steps before the program. But to use SixSigma at all, much prework is experiments start. Because this can be very time needed to build up structures which can handle SixSigma consuming it should be integrated in a project and to train green and black belts in this structure. This is management approach. very costly and time intense. Small and medium sized To be sure the DoE is used in combination with the enterprises (SME) do not have the resources for this. The system analysis and other tools the user will need a idea is not to do a complete SixSigma program, but to complete and simple guideline. structure DoE with DMAIC, which is the project management of SixSigma. This extensive way allows a Approaches to process improvement fast and target oriented proceeding to optimise processes.

There are different approaches, which are used for The five steps of fast forward process improvements. The most important ones are the PDCA cycle from Deming and the DMAIC of Six Sigma. This guideline will be taken from the SixSigma PDCA is build up by the four phases Plan, Do, Check, approach. DoE is one method of the SixSigma approach. Act. On the basis of a concrete problem the actual state SixSigma structures a project in five phases: Define, has to be defined and analysed. There is a wide range of Measure, Analyse, Improve and Control. Each of this methods and tools available for this operation /2/. phases involves different targets and tools. The approach First the causes of the problem have to be identified of SixSigma is strong target oriented. Fig. 4 shows the and then possible solutions have to be developed and structure of the “fast forward” approach prioritized (Plan). Afterwards concrete measures must be .

Define Definition of the objective, the project team and the duration

Regression Definition of the critical quality characteristics Calculation Measure Identification of critical influencing factors Optimum Search Analyse Designing experimental plan and operating experiments Screening Improve Experiments Controlling the fulfilment of the objective Control

Fig. 4: Approach for “fast forward”

Define: the problem is validated. For the optimisation the In the define phase the problem has to be identified. process/product will be analysed in detail. System and Additionally requirements are defined and the goal of the function analysis are made and the most important factors project is set. A team must be defined, that is able to and possible factor levels of the experiments are defined handle the considered process. Everyone who is related to /3/. the process has to be a member of this team /3/. In some cases it is possible to find simple improvements after the phases (D,M,A). But for an Measure: extensive optimisation, the next phases are the most In the measure phase the problem is validated and important ones. refined. With quality data and other available measures the problem is specified and refined /3/. Improve: In this step of the “fast forward” approach the critical In the improve phase ideas to remove root causes are quality characteristics of the chosen process/product, their developed. The solutions are tested and standardised. The measurement categories and qualified measurement results of the improve phase have to be measured and systems are defined. At this point the problems are documented. /3/ defined and the way of the measurement is described. “Fast forward” uses the phase “improve” to build up a plan for the experiments. Depending on the goal of the Analyse: project and the number of factors, different kinds of For the problem causal hypotheses are developed. design can be used. The improve phase can be repeated Then the root causes are identified and the hypothesis of several times, depending on the starting point of the 52 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 optimisation. The user can decide depending on the all DoE methods he wants to use. Fig. 5 shows some number of factors and other influences, which method of methods that can be used in the steps for optimisation.

Screening experiment Experimental search for optimum Examination in the area of optimum

• Factorial design [2 k-p] • Plackett - Burman • One-Factor-At-A-Time • Factorial designs [3k] Designs • Method of Steepest • Box-Behnken- • Group Screening Ascent Designs Designs • Simplex Design • Central Composite •... Designs • Evolutionary Operation •... • ... Reduction of variables Starting point for examination in the Regression equation area of optimum of the response

Fig. 5: Usable methods of DoE for different steps of optimisation. /2/

If there are many factors and the target is a regression was used in the project “ProSpray”, which is supported calculation, you start with screening experiments. After be the BMWA (German Ministry for Economics and analysing these experiments and finding a less number of Work). factors, an optimum search will be the next step. Analysing this optimum search will lead to a small The Project “ProSpray” number of factors. With these factors experiments for a regression analysis will be done. With this regression The project started in July 2003 and will end after analysis the process can be described in detail. But it is three years in July 2006. Members of the project not necessary to start each optimisation with screening “ProSpray” are the Materials Science Institute (WW) of experiments. Depending on the number of factors and the the RWTH Aachen, the Fraunhofer Institute for available process knowledge the improve phase can start Production (IPT) and the five industrial partners Zierhut with an optimum search or with an regression analysis. Messtechnik GmbH, GTV mbH, TLS GmbH, OBZ Additionally the improve phase depends on the intention GmbH and TACR GmbH. The opjective of the project is of the optimisation. If the objective is to find the most the development of an offline tool to control the process influencing factors, an optimisation can be conducted of thermal spraying. after the screening experiments. The process of thermal spraying is very complex and presently controlled by the experience of its respective Control: operator. It is a process with numerous process variables, To maintain the performance the standard measures numerous measurable responses and miscellaneous noise must be established. If necessary, existing problems must factors. The effects of the process variables on the be corrected. In this last phase “control” the reaching of responses are only rudimentarily known. Furthermore no the objective will be controlled. Process changing’s must process diagnostics which are easy to use and affordable be transferred to production and the success must be for small and medium-sized enterprises (SMEs), could be controlled /3/. established in the appliance of thermal spraying. This can be a starting point for a continuous In this paper the analyse of the APS procedure improvement process. (atmospheric plasma spraying) will be shown. Fig. 1 With this approach a process optimisation can be demonstrates how this process works. executed in a short time with best results. The approach

53 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Fig. 1: Atmospheric plasma spraying (1 – Plasma gas; 2 – Cooling water; 3 – Direct current; 4 – Coating material; 5 – Cathode; 6 – Anode; 7 – Substrate) [Gemeinschaft thermisches Spritzen e.V.

Between the cathode and the anode in the plasma gun an arc is build up with the help of a high voltage direct Usage of the “Fast Forward” approach with a current. With help of this arc and an inert gas (Argon) or thermal spraying process a mixture of Argon, Helium, Nitrogen or Hydrogen the plasma is build up. The coating material is pulverised and In the project “ProSpray” the “fast forward” approach flows with help of a carrier gas into the flame. In the was used to identify all significant main effects and flame of the plasma the coating material is melted and interactions, which affect the measurements. accelerated. After that the material hits the substrate and Starting with the define phase, the chosen process and sticks on it. an objective was defined. The intention was to find the There are a many important factors, that influence this most important factors that influence the measurement process. The measuring of the coatings is only possible results of the Pfi-system. In this case the used “black by destroying the sample. With the Pfi-System of Zierhut box” of the process contains two processes. One is the Messtechnik GmbH the process can be observed and general spraying process and the other is the measured during the coating. By making pictures of the measurement system of the Pfi-system. The team process the plasma stream and the particle stream are members are process responsibles of each company, approximated by ellipses. Fig. 2 represents this method. specialists from the Materials Science Institute of the RWTH Aachen and the Fraunhofer Institute for Production Technologies Aachen. In the next step the defined problem should be refined. For this the existing measurements are examined. So the prognosis was approved, that there are useful and repeatable measures with the Pfi-System, but the coherences between the factors and the Pfi-System are not known. In the step “Analyse” system- and function analysis were made. For the Aps process five factors were identified that should be investigated in the next step. • Flow rate of the first plasma gas (Argon) • Flow rate of the carrier gas • Amperage • Flow rate of the second plasma gas (Hydrogen) Fig. 2: Structure of the Pfi-System [Zierhut Messtechnik GmbH] • Powder rate (Flow of the coating material)

With these ellipses the process can be measured. Any Then it was worked out, if the coherences are linear or changes of the process will cause various positions and non linear. The experts suspect linear effect and measurements of the ellipses. But the kind of effects interactions. between the influencing factors and the measurements are In the improve phase, the experiments for DoE are not known. So the first step will be to find out, how the done. Because of the pre-work, the finding of an usable measurements are influenced by the factors like gas flow, design plan is simple. It must be a plan, that combines a powder rate and others. Later it must be worked out, how linear regression with main effects and interactions. the measurements are correlated with the coating Because no coatings are sprayed the number of parameters. experiments is not important. A full factorial design is chosen. With this, all interaction can be calculated. Additionally central point experiments are done to verify the acceptance of linear effects. A repetition of the experiments is necessary to be able to calculate the significance of the effects. After conducting the

54 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 experiments, the following results are identified: (Table CONCLUSION 1) “Fast Forward” enables especially small and medium Table 1: Significant effects and interactions. sized companies to use the powerful method DoE in an Ellipse x y a b alpha easy, fast and clearly structured way. The method of DoE right is complemented by an enclosing project management Corrected 0,95 0,9 0,95 0,96 0,95 tool. This contains the five phases Define, Measure, Analyse, Improve and Control of SixSigma. But it does stability not need the complete SixSigma structure with it´s index disadvantages of expensive trainings for black belts, Linearity no yes no yes yes green belts and its methods. It only uses the structure Argon (1) * * * * * DMAIC to arrange optimizing projects. With this, Hydrogen * * * * * process optimisation is more target oriented and can be (2) handled faster. The phases Define, Measure and Analyse make it possible to get as many information as possible Current * * * * - before planning the experiments. A repetition of the (3) phases Analyse and Improve to get from screening Carrier * * * * * experiments to regression analyses lead from the general gas knowledge of the process to an more detailed knowledge. flow(4) This makes it possible to use DoE in a more efficient and Powder - - - * * effective way. The Control phase will be used as an rate (5) approach for a continuous improvement. 1*4 - * - - * Using this approach less experiments are necessary, time and money is saved and companies will go “fast 4*5 - * - - - forward”. 1*2*5 - * - - * 1*3*5 - * - - - Legend * = - = not significant significant REFERENCES

For example the results for the right ellipse are /1/ Pfeifer, T.: Qualität in produzierenden Unternehmen shown. There are five measurements of the right ellipse. 2002 – Eine Untersuchung zum Zusammenhang zwischen The coordinates of the center (x, y), the length (a) and the Unternehmenserfolg und Qualitätsmanagement. Aachen: height (b) and the angle (alpha) of the ellipse. It is Fraunhofer-Institut für Produktionstechnologie IPT, 2002 /2/ Pfeifer, T.: Quality Management – Strategies, recognisable that all ellipse variables result significantly Methods, Techniques. München, Wien: Carl Hanser Verlag, of the main factors and their interactions. But in this case 2002 two problems have occurred. One is the nonlinearity of /3/ Pande, P.: The six sigma way – How GE, Motorola the x coordinate and the height of the right ellipse. The and other top companies are honing their performance. New problem are some significant three factor interactions, York: McGraw-Hill Companies, 2000 which are difficult to describe. But the important information for “ProSpray” was the information, that changings of the factors can be identified by variations of the measurements. To make the process adjustable, variances of the measurements must be corrected by suitable changings of the factors. Because the coherences are very complex and a regulation with usual methods is not practicable. The usage of neural networks will solve this problem. For neural networks the mathematical background is not important. They only need to learn the process characteristics. Then they are able to control the process. As shown in the example, the method “fast forward” can be used in different cases. In the case of “ProSpray” the procedure was used to the significant effects between the factors of the thermal spraying process and the Pfi- measurements.. Then the significant effects were worked out. Based on this results, the next steps were to find out the effects of the spraying process factors on the coating parameters and the correlations between pfi- measurements and coating parameters. With this basics a controlling with neural networks will be worked out.

55

Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

INTERNET BASED AUTOMATION OF THE PROCESS "DESIGNING - MACHINING" OF A WORKPIECE

Prof. Dr. K.-D. Bouzakis(1), Assistant Prof. Dr. A. Vakali(2), Lecturer Dr. G. Andreadis(1), E. Karapidakis(2) 1 Laboratory for Machine Tools and Manufacturing Engineering, Mechanical ; Eng. Dept., Aristoteles University of Thessaloniki, Greece; 2 Computer Science Dept., Aristoteles University of Thessaloniki, Greece

Summary: The purpose of this paper is to search the possibility of the use of XML (eXtensible Mark-up Language, the document processing standard proposed by the World Wide Web Consortium), in the description of machined parts (parts which are manufactured with machines as drills, mills, lathes etc). The objective is the independency of the part designing process from the fabrication process. Keywords: XML, CAD, CAPP, IGES, feature recognition, process planning, Electronic Data Interchange.

1. INTRODUCTION certain rules, and the constructor can manufacture the part. The purpose of this research is the independency of Therefore the use of a new language MP- ML the designing process of a part from the fabrication (Machining Process – Markup Language) for the process, via an interface between the designers and description of machined parts is proposed. The language fabricators of the parts. has designed so that it is suitable for the description of The process which is applied today is shown in the geometrical elements of a part, as much as for the following diagram (Fig. 1). determination of the workpiece’s process plan and the tools and machine-tools that will be used in the fabrication. The proposed system is constituted by two sub Design the Part Feature Extraction systems of software, DesignSite and ManufactorSite Software. The first one provides functionality for the production of MP-ML document, using data from IGES files, in order for the whole procedure to be independent from any specific CAD system. The second (ManufactorSite Software) provides functionality for the Processing with CAM Process Planning transformation of MP- ML document in G & M Codes for the control of the tool-machine.

2. THE MACHINING PROCESS MARKUP Machining NC machine and tool Data Base LANGUAGE (MP-ML)

The MP-ML is an application of XML /1/, the Fig. 1 Design and Manufacturing process document processing standard proposed by the World Wide Web Consortium (W3C), for the transfer of The designer draws a part with the aid of a CAD information about the machined parts between designers system. Then from this drawing the geometrical and constructors. structures are extracted and with the cross-correlation The format of MP-ML is defined in a Document with the suitable tools and tool-machines, the process Type Definition file (DTD) and thus the syntactic plan, for the manufacture of part is created. Then the correctness of the MP-ML documents, applying the DTD process plan supplies a CAM system, which produces as can be checked. The DTD also can be stored in Internet, final product the program for the tool-machine, which so that it is accessible by every user. finally will manufacture the part. The XML is derived from SGML (Standard The above process is time-consuming and prone in Generalized Markup Language) and was drawn in order errors, and certain, always exist the problem of to make the data “machine readable”, in contrast to compatibility from system to system. HTML which has constant and concrete number of tags. It is essential therefore the creation of a mechanism XML does not have concrete tags but the user must (interface), which will separate the work of designing define his own tags, adapting thus, the XML to his from the work of manufacture, with an independent from needs. A large number of XML /2/ applications have individual systems way. So the designer will know that been developed for various needs, as MathML for the his drawing is constructible as long as he has observed

57 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 definition of mathematical equations, SML for the steel The MP-ML document contains all information that is industry, etc. essential for the description and machining process of a machined part. This information, as appear in figure 2, is Advantages of the XML (thus MP-ML) documents found under the root element “part”, which contains five are: elements. • The XML based documents can be read with the aid These elements are: of a simple Web Browser such as Internet Explorer a) element Header: it contains general information that 6.0, without any change and without any specific concerns the machined part, which is defined in the software. document, as well as information for the designer of the • There are a lot of tools for their handling and their part. publication in Internet. b) element Rawmat: it contains the description of the • There are two languages for formatting the XML initial volume of the raw material, from which with the documents: CSS (Cascading Style Sheet) and XSL machining processes the final workpiece will be (eXtensible Style Language), with which the manufactured. The description of the workpiece volume appearance of the XML documents according to the includes the consisting material, its shape (cube “CUBE”, needs at any time, can be modified. cylinder “CYLINDER” or cutting surfaces “SURF”) and • XML is extensible. the coordinates which delimit the initial volume. • Finally there is an abundance of libraries in most c) element Features: it contains the geometrical and programming languages as C++ and Java for the topological structures (features) as they have been XML documents processing. extracted. These geometrical structures should be removed from the initial volume with the appropriate Due to the above mentioned advantages, XML was machining processes in order to manufacture the part. selected to be the base for the MP-ML creation. The Each geometrical structure is defined by the surfaces Document Type Definition (DTD) of MP-ML is which delimit it. presented in the next section. d) element processPlan: it contains the process sequence (process plan) needed for the manufacturing of the workpiece. e) element tools: It contains the cutting tools which will 3. THE DOCUMENT TYPE DEFINITION be used at the operations that are described in FOR MP-ML processPlan. It is constituted of one or more tool elements. 4. System Software The following figure (Fig. 2), shows the simplified tree In order to be able to check the function of MP- ML, two structure of MP-ML documents software applications were created. The first one in the designer site (DesignSite), and the other one in the manufacturer site (ManufatorSite) (Fig. 3).

CAD file

Web compatible Graphic model IGES file (VRML)

*.dat files MP-ML

DesignSite

Internet ManufactorSite

MP-ML

G & M Codes

Fig. 3 System process flow Fig. 2 MP-ML Structure

58 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

4.1 DesignSite: 4.2 ManufactorSite: At the designer site the following operations must be In the manufacturer site the following operations must considered: be supported: • The drawing of the part • The recognition of the part that has been described in • The recognition of the geometrical structures, that MP-ML document should be removed from the initial work piece in • The creation of a better process plan for the order to manufacture the part machining of the part • The creation of the process plan • The production of the G&M Codes file that will lead • The production of the MP- ML document the NC machine to manufacture the part. • The production of the VRML /3/ model ManufactorSite Software is responsible to import the In the frames of this work there was an effort for MP-ML document in the constructors CAM system. using existing software, where this was feasible. Thus, for The development of such a translator–generator the designing, the usage of any CAD software is possible, would make the whole procedure dependent from the due to the fact that the whole procedure is depending in particular NC machine. In order to avoid such a the ability of CAD systems to export IGES files. For the dependency, the development of a CAM Importer, in the feature recognition and the process plan creation, site of the manufacturer, was preferred. This utility software which has already been developed in a previous recognizes the described in MP-ML document part and work /4/, was used. This software uses IGES files as input then imports it in the constructors CAM, which then with and it produces ASCII files, (*.dat files in figure 3), that the usual process manufactures the part . define the geometrical structures (features) and the The selection of the CAM importer development, procession of work (process plan) for the construction of the instead of a G & M Codes Generator appears preferable, part. since this development is easier and does not limit the At the same time also a “dat to MP- ML converter” for constructor in ManufactorSite Software abilities; however the transformation of dat files into MP-ML documents, was it provides the possibility of using any preferable CAM developed, as well as a “VRML Generator” for the VRML system. model creation from MP- ML document. The creation of the VRML model was preferred, in order for the part to be easily presentable in a WEB browser without any special software. 5. AN APPLICATION EXAMPLE In this work context, the main interest is how much, such a schema as MP-ML is able to transfer all the necessary An application example is examined in this section. information, for the fabrication of a part, from the designer The part in figure 4 has been drawn with a to the manufacturer. commercial CAD System and it will constitute an The MP-ML document in this stage contains all the example for the methodology presented in this paper essential information for the part fabrication and is dispatched via Internet to the constructor as shown in figure 3.

. Fig. 4 An application example (VRML file shown in a web browser)

59 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

First the part was designed and then the CAD data REFERENCES was exported in an IGES file, which after that was used as input for the feature recognition process and the /1/ Eckstein, R., 1999, “XML Pocket Reference”, process planning. O’Reilly More specifically, the following files were produced: /2/ Michael Morrison, et al. , 2000, “XML Unleashed”, • Perim.dat file: It contains all the surfaces of the Sams Publishing. part. Each surface is defined by the edges that delimit it, /3/ VRML, ISO/IEC 14772-1, http://www.vrml.org while each edge is defined by the coordinates of the /4/ K.-D. Bouzakis, G. Andreadis : An integrated process starting point and the ending point of it. There is an planning system based on feature recognition and machinability criteria, International Journal for Manufacturing Science & exception with the edges that are arcs of a circle. For Technology, Vol. 3, No. 2, (2001), pp. 80-91. these edges the coordinates of the centre point of the circle, its radius, as well as the angle of the circular sector that corresponds in the arc, are recorded in addition. • Rawmat.dat file: It contains the surfaces that shape the initial raw material volume of the workpiece. As initial volume is defined, the volume of the crude piece from which the final part will result, after the machining processes. This volume results from the minimum bounding box that includes the machined part. • Feature.dat file: It contains all the topological entities of the part. The description of each entity becomes with the recording of surfaces that delimits the volume that should be removed from the work piece in order to construct the entity. • processPlan.dat file: In this file, the required setups and the operations for the fabrication of the part are described. The above files were translated into an MP-ML document and the equivalent VRML model that appears in the picture (Fig. 4) in Internet Explorer 6.0.

6. CONCLUSIONS AND FUTURE WORK

This work presents the MP-ML language, a language for the description of machined parts, which is based on XML and is compatible with web-based applications. A software system using the MP-ML language was presented also. This system can start from a CAD design of a part and leads to the machined part. More specifically, the system which was presented aims at the independence of the designing process of machining parts from the manufacturing process, so that the part design can be accomplished in a different place that would abstain far away from the manufacture place. The MP-ML as an XML based application is from his definition Internet compatible. Also the procedures of feature recognition of geometrical structures and process planning are based in IGES standard and consequently are independent from any CAD system that is used for the designing of the part. Finally the system is modular thus easily extensible, providing with particularly low cost and low equipment requirements. A work environment even in one simple PC can decrease the total cost and manufacture time of the part. Certainly MP-ML is in a very precocious stage of development and more systematic research is needed in order to extend it where it is needed.

60 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

DEVELOPMENT OF AN INTELLIGENT MODULE FOR DECREASE OF MEASURING ERROR ON CMM

M. Stevic1, J. Hodolic1 S. Vukmirovic2 1Insitute for Production Engineering, Faculty of Engineering, University of Novi Sad, Novi Sad, Serbia and Montenegro; 2Department of Control Systems and Automation, Faculty of Engineering, University of Novi Sad, Novi Sad, Serbia and Montenegro

Summary: The first segment of the paper proposes an overall model for the assessment of the correction factor of measuring error.The second segment of the paper proposes developed software solution for the assessment of the correction factor of measuring error on coordinate measuring machine (CMM).The paper ends with some final remarks. Keywords: CMM, measuring error, neural network

1 INTRODUCTION in transportation, and costs. Yet, the results of assessing the quality parameters on CMM and the accuracy in Coordinate measuring machines (CMM) have been measuring, whatever the technique used for obtaining largely applied in the industrial and laboratory conditions them, can be utilized for compensating the measuring after they gave revolutionary results in monitoring the error. process of measuring, inspection and quality control in general. Among the number of properties of CMM there are four key ones that place these systems onto the highest level categories when compared to other small- 2 INFLUENCING FACTORS AND and medium-series production. These properties are as CORRECTION FACTOR OF MEASURING follows: universality, flexibility, productivity, and ERROR ON CMM accuracy. Calibration, verification, and compensation of the When analyzing the factors influencing the measuring measuring error have been emphasised as important tasks result with the application on CMM, it is easy to for maintaining high accuracy. conclude that these are numerous and heterogeneous. In last two decades, many methods for determining These influencing factors can be presented as in figure 1 the measuring accuracy of a measuring instrument in a /2/. measurement space have been proposed and applied. Determining correction factor of the measuring results Basically, these methods can be classified into three from the point of mathematic correction functions is groups of standard techniques; these are /1/: possible to perform in two manners /3/: • kinematic reference standard technique, 1. Selecting one of the existing correction functions • parametric calibration technique, based on the entered relevant data for correction, and • transfer standard technique. 2. Inputting relevant data obtained by the monitoring system and by choice, or calculating the correction Characteristic example of the kinematic referential functions. standard technique is a ball bar. This technique is very simple for obtaining data; however, it is difficult to cover The manner of determining correction factor to be the whole measurement space of the machine. used depends on the source and the type of measuring Parameter calibration technique serves to acquire error. After analyzing basic factors influencing the information for diagnosing the error on the machine. Yet, measuring result and grouping them according to mutual this technique lasts long, requires expensive equipment connections and influences, figure 2 can be formed as a and specially trained operator for the equipment, e.g. for premise for creating software solution for determining laser measuring system. correction factor of measuring error. Transfer standard technique is limited due to the Generally speaking, two schemes of compensating impossibility to produce machine tools of small weight, measuring errors are possible: subsequent compensational high thermal stability, small costs, etc. Besides, different scheme and real-time compensational scheme. With sizes of standard elements are required for different subsequent compensational scheme, compensation is dimensions of measurement space on machines, and there formed after measuring procedure is finished. Real-time can also appear some problems concerning storage, compensational scheme is when the compensation is handling and transportation. formed in real-time base, that is, when the measuring Unfortunately, none of the existing techniques does procedure is in progress; this scheme has the integrative not fulfil all the demands in connection to working time, map of spatial errors /3/. usage simplicity, error diagnostics, thermal stability, ease

61 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Device axes Temperature Type and value Temperature of form deviation Control M e fluctuacion Length expansion a Compressed air s E coefficient u Temperature gradient W n r i v o n i Microstructure, r r g Temperature radiation o k Length measurement p i n roughness,cracks n i m e system s Humidity c t en r e Probing system u Elasticity m Vibrations t e Plasticity Evaluation software n Dirt t Mass CAA Measurement result

Education Probe selection Experience Usage of a forth device axis Care Probe change Integrity Clamping Measuring sequence Filter planning Number and distribution of points Measurement method

Figure 1 Factors influencing the result of measuring on CMM /2/

Fast development of personal computers (PC), as of this problem led to a general model for determining well as various accurate and available hardware any correction factor, whether or not a quality equipments, has led to the fact that, in several last years, parameter. This model contains four basic steps: measuring systems based on PCs have taken the  measuring preparation primacy over expensive and specialized equipment. User makes a test plan of influencing factor for Together with lower costs and faster profitability, there measuring accuracy. is also higher productivity, better process quality, higher  measuring or monitoring process accuracy, reduced number of errors, etc. These Measuring or monitoring gives genuine measuring facts enable that, instead of mathematical functions for values. Measured results are stored in appropriate correcting measuring error, integrative map of spatial databases. errors can be used, or the value of the error determined  results processing for the given measuring point. On the basis of measuring result processing, measuring correction factor is determined. If observed influencing factor is a quality parameter, it is also evaluated.  output results. User chooses the way for presenting obtained results.

These steps can be presented by a general model for determining correction factors of measuring error, whether or not caused by influencing factor as a quality parameter (figure 3). Figure 2. Mutual connections of influencing factors on measuring result /3/

3 CORRECTION FACTOR OF MEASURING ERROR ON CMM

CMM presents complex products that have their own quality parameters. Quality parameters and other influencing factors affect the accuracy of measuring on CMM. Increasing measuring accuracy by compensation of measuring error based on correction factor is a way to increase the accuracy of measuring on CMM. Analysis

62 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

4 PROPOSITION FOR SOFTWARE SOLUTION FOR DETERMINING CORRECTION FACTOR OF MEASURING ERROR

Following the analysis of factors that influence the measuring result while measuring on CMM and their mutual connections, the concept of the software solution for correcting measuring error is proposed and presented in figure 4. Software solution has been formed on the principle of modular building with establishing necessary connections between certain modules. The solution has 5 basic modules: MI-measuring instrument, OK-environment, MP-workpiece, OP-operator, MS- measurement strategy.

MI-measuring instrument OK-environment 1-Device axes 1-Temperature 2-Control 2-Temp. fluctuac. 3-Compressed air 3-Temp. gradient 4-Lenght measurement system 4-Temp. radiation 5-Probing system 5-Humidity 6-Evaluation software 6-Vibrations 7-CAA 7-Dirt MP-workpiece OP-operator 1-Type and value of form deviat.1-Education 2-Lenght expansion coefficient 2-Experience 3-Microstructure 3-Care 4-Elasticity 4-Integrity 5-Plasticity 5-Measuring seq. 6-Mass planning MS-Measurement strategy 1-Probe selection 5-Filter 2-Usage of a forth axis 6-Number-and distri- 3-Probe change bution of points 4-Clamping 7-Measur. method.

Figure 3. General model for determining correction factor of Figure 4. Concept for software solution for the correction of measuring error /4/ measuring error while measuring on CMM /3/

63 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Moving within the software solution on the level of one module is possible through adequate levels presented in figure 5.

Figure 5. Levels within one module of the software solution

Calculating correction values within particular With the existing CMM, static sources of errors can modules depends on the source and the type of error, and usually be compensated by additional software. The it can be performed by the following methods: correction of dynamic errors with the existing machines  algebraic method is extremely difficult. Therefore it is important to Total error, which is also the correction factor for one optimize the response of mechanical structure to dynamic axis, is determined by algebraic addition or subtraction disturbances. It is possible to do it even before finishing (depending on the direction of error produced by the prototype by using Lagrange's energetic method and influencing factor) of individual error components. This checking the results by modal analysis, like in /8/. method is very usable, and characteristic example in  finite elements method literature is /5/. When analyzing the compensation of measuring error  matrix method produced by dynamic influences, the finite elements It is based on presenting measuring error by matrix method is often used, especially in the cases of fast equation. It is mostly used when measuring is done in measuring, like in /9/. different coordinating systems for traversing from one to  inversion task method another system with the help of transformation matrices, As current researches have shown for nonstationary like in /6/. The method is practically reduced to algebraic. and nonlinear processes in technology, researching is  linear interpolation method successfully done by a new scientific approach based on It is applied when the measuring error appears the inversion task method. This method can be applied because of an influencing factor in characteristic dots of when there is a problem of accuracy on CMM in the part measuring line. The error in remaining dots is determined of kinematic accuracy and thermal changes and their by linear interpolation between characteristic dots, e.g./7/. influence on the measuring accuracy, as can be seen in  smallest square method /10/. It is used for determining compensation function of  genetic algorithm method measuring error by the results of measuring error in Genetic algorithms are heuristic optimizing methods characteristic dots, e.g. /7/. By the smallest square imagined as an imitation of natural evolution. Genetic method mathematical function which best describes the algorithms imitate natural evolution in such a way that change of measuring error along measuring line is being the optimizing process presents the environment in which searched for. individuals – input process data – live. Problems of  energetic equation method 64 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 measuring error can be analyzed using this method, as in be noticed that there is a part where results are /11/. completely satisfactory. Figure 7 presents comparative  neural network method results gained by neural network for the third year and by Human associative judgement and recognition is LMS for position error X axis. based upon comparing (associating) previously experienced and remembered situations to currently active ones, Similarity is using as a basis and leads to a conclusion already succesfuly proven in previous situations. Artificial neural networks are capable to function in a certain way like principles of associative judgement and recognition. Neural network is a system made of several simple processors (units, neurons); each of them has a local memory for remembering processing data. These units are connected by communication channels (connections). Data exchanged in this manner are usually numerical. Units process just their local data and inputs they receive via connection. Local operator's restrictions can be removed during training. Neural network has wide and heterogeneous application in the field of CMM accuracy, like in /12/. Neural networks can be employed to calculate, recognize, and predict the values of correction factors. Special problem is assessing the value of a correction Figure 7. Comparative results of real value of measuring error factor as a part of the intelligent module for determining and expected values obtained by neural network training correction factor. In order to perform assessment, the change of a It is realistic to assume that for a larger training set of correction factor for a specific machine during a time measuring error changing in time and for a certain period has been monitored. In this case, based on known machine, the results could have been much better. inputs of correction factor from the periods of its Neural network is trained by propagation of error determining, the network has been trained to assess its backwards, and the activating functions tansig and value in the future. Configuration of neural network purelin have been applied (figure 8). applied for assessing correction factor of measuring error is shown in figure 6.

Figure 6. Configuration of neural network applied for assessing correction factor of measuring error Figure 8. Activation functions tansig and purelin The network currently has two inputs, and it is trained for the period of one year. The number of inputs can be The network can be trained on the same principle for easily changed, so for some following year we can have any complete module shown in figure 3, for a part of a more inputs. In that case, our network will be even more module or for several modules simultaneously. trained to solve this problem. Unfortunately, a small number of available training results disabled the possibility of output results to be qualitative enough for the change. However, on the comparative diagram of real and expected values, it can

65 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

5 CONCLUSIONS /10/ Barakat, N.-A., Elbestawi, M.-A., Spence, A.-D., 2000, Kinematic and geometric error compensation of a coordinate measuring machine, International Journal of During the analysis of the sources of measuring errors Machine Tools & Manufacture, 40:833-850. on CMM it has been determined that the factors /11/ Wen, X., Song, A., 2003, An improved genetic influencing the measuring results are numerous and algorithm for planar and spatial straightness error evaluation, heterogeneous. Errors originating from these factors can International Journal of Machine Tools & Manufacture, be more or less successfully compensated by determining 43:1157-1162. correction factor of measuring result. The paper presents /12/ Dong, C., Zhang, C., Wang, B., 2002, Prediction and the general model for determining correction factor of Compensation of Dynamic Errors for Coordinate Measuring measuring error originating due to an influencing factor. Machines, Journal of Manufacturing Science and Engineering, 124:509-514. The paper specially emphasises the modular building of the software for determining correction factor of measuring error and the possibility of predicting the change of measuring error in time by using neural network. The results obtained by neural networks would be better if there were more available results for training. When one summarizes the activities encircled in this paper, it can be said that this is another step towards increasing the measuring accuracy of CMM.

REFRENCES

/1/ Silva, A.J.B., Burdekin, M.: A modular space frame for assessing the performance of co-ordinate measuring machines (CMMs), Precision Engineering Journal of the International Societies for Precision Engineering and Nanotechnology, No26,2002. /2/ Weckenmann, A.: Influences on result value and uncertainty using flexible measuring instruments, Summer Academy 2001., Dubrovnik 2001. /3/ Stevic, M., Hodolic, J., Gatalo, R., 2002, Development of measurement results correction method on numerically controlled measuring machines (CMM), 29. International conference of production engineering, Proceedings, Belgrad. (in Serbian) /4/ Stevic, M., Hodolic, J., Vukmirovic, S. , 2004, Development model for correction of measuring error on CMM, 11. International CIRP Life Cycle Engineering Seminar «Product Life Cycle – Quality Management Issues», Proceedings, 217-222, Belgrad. /5/ Ačko, B., 1997, Method for analyzing measure accuracy of Coordinate measuring machines, Ph.D Thesis, University of Maribor, Faculty of Mechanical Engineering, Maribor. (in Slovenian) /6/ Pahk, H., Lee, K., Lee, J., 1996, Real time compensation system for errors introduced by measurement probe and machine geometry in comercial CMMs, The Fourth International conference on automation Technology, Proceedings 1:309-315, Taiwan. /7/ Stevic, M., Hodolic, J., 2001, Testing and correction of Coordinate measuring machines quality parameters, International scientific-technical conference mechanical and technique on start XXI century, Proceedings, 3:227-231, Doneck. (in Serbian) /8/ Nijs, J.-F.-C., Lamers, M.-G.-M., Schellekens, P.-H.- J., Wolf, A.-C.-H., 1988, Modelling of a Coordinate Measuring Machine for Analysis of Its Dynamic Behaviour, , Annals of the CIRP, 37/1:507-510. /9/ Mu, 1999., Dynamic Error Compensation of Coordinate Measuring Machines for High-Speed Measurement, The International Journal of Advanced Manufacturing Technology, 15:810-814.

66 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic

TRACEABILITY OF MEASUREMENT RESULTS IN INDUSTRY

B. Ačko University of Maribor, Faculty of Mechanical Engineering, Maribor, Slovenia

Summary: Many decisions about the control of manufacturing processes depend on measurement results of different quantities as length, pressure, temperature, mass, electrical quantities etc.. Unfortunately, not enough attention is paid to the "quality" of measurement results in industry. This was the main reason for us to write this article, which will introduce the most important aspects of traceability of measurement and uncertainty of measurement results. Expressions "traceability" and "uncertainty" reflect the quality of measurement results and therefore our trust in the results. Different metrology committees, working groups and councils in Slovenia work hard on establishing metrology system that will be compatible to European system an will offer the industry the possibility to improve the quality of measurements. This system is based on national standards of measurement, which are also described in this article. Keywords: Traceability, Calibration, Uncertainty of Measurement, Measurement Result, National Standard

1. INTRODUCTION not eliminated are temperature deviations and deviations caused by measurement equipment. Al systems for controlling manufacturing process are based on measurement or inspection results. These results are usually used as input quantities for statistical 3. CALIBRATION OF MEASUREMENT evaluation of manufacturing process. Therefore, the EQUIPMENT quality of statistical evaluation directly depends on the quality of measurement result, which is determined by proper accuracy (expressed as uncertainty of 3.1 Definition measurement), reliability, proper form and availability at defined time. In order to perform quality measurements Calibration means determining and documenting the and to produce quality measurement results, it is deviation of the indication of a measuring instrument (or necessary to choose well educated and experienced the stated value of a material measure) from the personnel, adequate equipment and measurement conventional true value of the measurand [4]. procedures, as well as proper environmental conditions. The conventional true value is represented by the Al the influences on measurement results should be measure of a standard or measurement device of higher known and controlled. accuracy level with a metrological quality already determined by calibration with a higher-level standard. Calibration hierarchy will be discussed in later chapters treating traceability. 2. INFLUENCES ON MEASUREMENT RESULT IN DIMENSIONAL METROLOGY 3.2 Reasons for calibration No measuring instrument or standard of measurement Influences on measurement results can be divided into is absolutely accurate. Even new instruments or standards four basic groups [2,6]: show some deviations from nominal value. This • operator, deviations change during certain time period because of • measuring instrument, the drift and the conditions of use. If we want to perform • measurement method and procedure, quality measurements, it is necessary to evaluate • environmental conditions. deviations by means of calibration and to eliminate them Each group is divided into systematic and random from measurement results [1,2,6]. influences [1,2, 3]. Systematic influences can usually be Reasons for calibrations can also be requirements of well defined and eliminated from measurement result. the standards for quality systems (e.g. ISO 9000) or in Random influences can be partially eliminated by special cases requirements of state regulations. statistical evaluation if it is possible to make enough measurements. In many cases statistical evaluation is not 3.3 Calibration period possible and measurement results contain random Calibration period depends on the type of the deviations, which decrease their reliability. Reliability instrument (or standard), frequency of operation and the can also be decreased by not evaluating and eliminating conditions of use. The period should be determined by systematic influences. Unfortunately, this often happens the user of the instrument. It is based on technical in industrial praxis. The most critical influences that are

67 Third International Working Conference “ Total Quality Management - Advanced and Intelligent Approaches “ knowledge, experience, history of the instrument, and 4. TRACEABILITY OF MEASUREMENT recommendation of experts. The period should be defined separately for each instrument in use. The term traceability means a process whereby the indication of a measuring instrument can be compared, in 3.4 Calibration certificate one or more stages, with a national or international Calibration results are given in a calibration standard for the measurand in question. certificate or a calibration report, which should contain at In each of these stages, a calibration has been least the following information: performed using a standard with a metrological quality • title “Calibration Certificate (Report)”, already determined by calibration with a higher-level • name and address of the performer of calibration standard [1,6]. Calibration hierarchy for length standards (laboratory, company), in Slovenia with indicated national standard and its traceability to international level (primary standard) is • unique number of the certificate, shown in Figure 1. • date of calibration and issue date, • page number and total numbers of pages on each page, 4.1 Importance of traceability • data about customer, Traceability of measuring and test equipment to • data about the measuring device or the standard national standards by means of calibration is necessitated being calibrated, by the growing national and international demand that • environmental conditions, manufactured parts be interchangeable. Supplier • measurement res. calibration method, companies that produce products, and customers who • general statement about the traceability of install them with other parts, must measure with “the measurement results, same measure”. In other words: “Measurements results • way of expressing uncertainty of measurement, must be comparable on international level”. • measurement (calibration) result There are legal as well as technical reasons for • signature of the performer of calibration, traceability of measurement. Relevant laws and • signature of the person approving the calibration regulations have to be complied with just as much as the results and a seal, contractual provisions agreed with the purchaser of the product (guarantee of product quality) and the obligation • uncertainty of measurement, to put into circulation only products whose safety, if they • standards and measurement equipment used for are used properly, is not affected by defects. If binding calibration. requirements for the accuracy of measuring and test Good certificate usually also contains remarks (e. g. equipment have been stipulated, failure to meet these injuries of the calibrated item). Date of the next requirements means the absence of a warranted quality calibration can be marked on the certificate only as a with considerable consequent liability. remark and not as a suggestion to the customer, because If it becomes necessary to prove absence of liability, it is always up to customer to define calibration period. the producer must be able to demonstrate, by reference to If the certificate does not contain all the above stated a systematic and fully documented system, that adequate items, it is not complete and should be refused by the measuring and test equipment was chosen, was in proper customer as not appropriate. Special attention should be working order and was used correctly for controlling the paid to measurement (calibration) results because many product. calibration institutions still issue certificates without There are similar technical and legal reasons why results. Many certificates also do not state uncertainty of calibration and testing laboratory operators should have measurement or the way of expressing uncertainty. consistent control of measuring and test equipment in the manner described. 3.5 Performers of calibrations Calibrations can be performed by national 4.2 Elements of traceability laboratories, by calibration laboratories, which somehow Traceability is characterised by a number of essential prove their competence for certain types of calibrations elements [1,6,7]: or by users of measurement devices. Competence for performing calibrations is proved by accreditation • an unbroken chain of comparisons going back to according to the standard EN 45001. This accreditation is a standard acceptable to the parties, usually national or valid if it is performed by national accreditation body or international standard; one of the EA (European Co-operation for Accreditation) • measurement uncertainty; the measurement members [1, 4]. uncertainty for each step in the traceability chain must be calculated according to agreed methods and must be stated so that an overall uncertainty for the whole chain may be calculated; • documentation; each step in the chain must be performed according to documented and generally acknowledged procedures; the results must equally be documented;

68 Third International Working Conference “ Total Quality Management - Advanced and Intelligent Approaches “

evidence for their technical competence, e.g. by International level demonstrating that they are accredited; • reference to SI units; the chain of comparisons National must end at primary standards for the realisation of the SI laboratory units; • re-calibrations; calibrations must be repeated at appropriate intervals; the length of these intervals will depend on a number of variables, e.g. uncertainty Accredited cal. Accredited cal. required, frequency of use, way of use, stability of Laboratory 1 Laboratory 2 equipment In many fields, reference materials take the position of physical reference standards. It is equally important that such reference materials are traceable to relevant SI In-house calibr. units. Certification of reference materials is a method that is often used to demonstrate traceability to SI units Company 1 Company 2 Company 3 Company 4 [1,2,7].

Figure 1: Different ways of assuring traceability of production measurements

• competence; the laboratories or bodies performing one or more steps in the chain must supply

expanded uncertainty calibration procedure of measurement primary standard (frequency 0,02 μm + 0,2⋅10-6⋅L absolute interferometric stabilized laser) 0,03 μm + 0,3⋅10-6⋅L measurement L-length of a gauge block k=2 level 1 mechanical comparison slovenian national standard- gauge block calibrated by gauge block comparator standard level interferometry -6 calibration class K or 0,05 μm +0,5⋅10 ⋅L quality class 00 k=2 level 2

standard- gauge block calibrated by mechanical comparison -6 gauge blocks on level 2 0,1 μm + 1⋅10 ⋅L gauge block comparator k=2 calibration class K or quality class 00 level 3

standard- gauge block standard- gauge block mech. comparison calibrated by gauge calibrated by gauge 0,2 μm + 1⋅10-6⋅L gauge block blocks on level 3 blocks on level 3 k=2 comparator quality class 0 res. 1 quality class 0 res. 1 level 4

Figure 2: Calibration hierarchy for length standards in Slovenia

At the international level, decisions concerning the 5. CALIBRATION HIERARCHY International System of Units (SI) and the realisation of the primary standards are taken by the Conference 5.1 International level Generale des Poids et Mesures (CGPM). The Bureau International des Poids et Mesures (BIPM) is in charge with co-ordinating the development and maintenance of 69 Third International Working Conference “ Total Quality Management - Advanced and Intelligent Approaches “ primary standards and organises intercomparisons on the calibration laboratory or national laboratory. The in- highest level. house calibration may be evidenced by a factory calibration certificate, a calibration label, or some other 5.2 National Metrology Institutes suitable method. The calibration data must be retained for a prescribed period of time [6,7]. The national metrology institutes are the highest authorities in metrology in all countries. In most cases they maintain the national standards of the country, which are the sources of traceability for the associated 6. NATIONAL STANDARDS IN SLOVENIA physical quantity in that country. If the national metrology institute has facilities to realise the The policy of establishing national standards in corresponding SI unit of measurement, the national Slovenia differs a little bit from common policy of standard is identical to or directly traceable to the primary national metrology institutes described in 5.2. In Slovenia standard realising the unit. National metrology institutes exits no national metrology institute, which would are responsible for disseminating the units of maintain national standards for all quantities. In fact, measurement to users who may be scientists, public national standards for different quantities are maintained authorities, laboratories or industrial enterprises and are by different laboratories located at institutes and therefore the top level of the calibration hierarchy in a universities. All the laboratories maintaining national country. The European National Metrology institutes standards were authorised by the state (MIRS). The basic (also laboratories from Slovenia) co-operate in condition for authorisation was international EUROMET. accreditation according to EN 4500. At the moment there exist national standards for mass, length, temperature, 5.3 Accredited calibration laboratories time and frequency, and current. National standards are made traceable to primary Accreditation is official recognition of competence. standards by calibration at international level in different Accreditation bodies accredit laboratories in industry and European national laboratories. other organisations, according to well-established criteria [6]. These criteria are laid down in ISO/IEC 17025. For 6.1 Requirements for national standards some topics EA has developed application publications, Accreditation is generally given for specified measurands National standards in Slovenia were established in for the smallest uncertainties that can be achieved with accordance with the following requirements: the suitable measuring devices available in the respective • best measurement capability of the laboratory calibration laboratory (“best measurement capability”). maintaining national standard should be at such level that Accredited laboratories are often at the top of a firm’s all demands in industry regarding required accuracy internal calibration hierarchy. Their task is to compare, at could be fulfilled, appropriate intervals, the firm’s own working standards • needs for calibration abroad should be at with reference standards, which are calibrated by a minimum level, national metrology institute or an accredited laboratory • costs for establishing and maintaining standards with a suitable best measurement capability. should be at such level, that the industry can effort it. Many accredited laboratories carry out calibration for Concerning these requirements, the most appropriate third parties on request, e.g. for companies that do not level of national standards in accordance with Fig. 2 is have calibration and measurement facilities with suitable level 2. Level 1 standards (primary) would be too equipment, and for private test laboratories working in expensive for the industry and there is no need for having the field of product certification. them, because the demands for calibration of If an accredited laboratory is to be engaged to carry measurement equipment and standards on the top out particular calibration, the client must ensure that the accuracy level are very small. measurement uncertainty achieved is appropriate for the intended use of the calibrated instrument. 6.1 Introduction of the standard The European accreditation bodies for calibration National standard of length was established in laboratories co-operate in the European co-operation for October 1998 in the Laboratory for Production Accreditation (EA). The goal of EA is to ensure that Measurement located at the University of Maribor – calibration certificates, issued by any accredited Faculty of Mechanical Engineering. It is realised by three laboratory, are accepted in the other countries. This sets of gauge blocks: implies that the laboratories and the accreditation bodies have to operate in a compatible way. • set 0,5 mm to 100 mm (122 pieces, K. Frank, class 00), 5.4 In-house calibration (factory calibration) • set 125 mm to 500 mm (8 pieces, KOBA, class K), An in-house calibration system ensures that all • set 600 mm to 1000 mm (5 pieces, Tesa, class measuring and test equipment used in a company is K). calibrated regularly against its own reference standards. For the purpose of maintaining and the use of the The company reference standards shall have traceability national standard, the laboratory was fully rebuilt and of measurement by being calibrated at an accredited special microclimatic chamber was integrated in one of

70 Third International Working Conference “ Total Quality Management - Advanced and Intelligent Approaches “ the laboratory rooms. The personnel consists of four well 6.3 Environmental conditions educated and experienced people working on calibration, research, development of procedures, and the design and The microclimatic chamber of dimensions 5 m X 2,5 construction of measurement devices. Calibrations are m X 3 m assures very good conditions regarding performed by mechanical comparison. temperature, humidity and cleanness. The temperature gradient in the calibration place (about 800 mm above the floor level) is not grater than 0,1 K. Temperature is 6.2 Measurement equipment measured with the system Zeiss TEMP 10 including 4 The equipment for transferring measures from the contact and two air sensors. The system is being standard consists of two comparators: calibrated periodically with an uncertainty of 10 mK and • standard gauge block comparator Mahr 826 for shows very good long term stability (the drift of all measures up to 100 mm, sensors was in the limits of ±10 mK in three years). • standard gauge block comparator Cary When calibrating short gauge blocks, the temperature is • comparator for gauge blocks up to 1000 mm, measured on the table of the comparator, while the designed, built and tested by the laboratory staff (Figure temperature during calibration of long gauge blocks is 3) measured in two points on the reference standard and in The comparator for gauge blocks up to 1000 mm was two points on the gauge block being calibrated. built on the existing 1-D machine Zeiss ULM 3000. Deviations of gauge blocks are measured with inductive 6.4 Uncertainty of measurement probes Mahr and the results of calibration are processed Calibrations are currently performed with the by our own software. following uncertainties:

• Calibration of short gauge blocks (0,5 mm – 100 mm) with u = 0,05 μm +0,5⋅10-6⋅L • Calibration of long gauge blocks (125 mm – 1000 mm) with u = 0,1 μm +1⋅10-6⋅L Calibration with such uncertainty can cover calibration needs even for the most precise industry at the moment and in the future. There are neither economical nor technical reasons for involving more precise national standard.

6.5 Accreditation The laboratory was accredited in accordance with EN 45001 (in 1997) and after that according to ISO/IEC 17025 for calibrations of short and long gauge blocks and gauge block comparators (short and long) with the above stated uncertainties by Slovenian accreditation body SA and by RvA from The Netherlands.

6.6 Financing The maintenance of the national standard is financed by the Ministry of Higher Education, Science and Technology. Purchase and development of new equipment depends on the projects (industrial, national and international) the laboratory is involved in. Only actual work on calibration is financed from the invoices issued to the clients.

6.7 Development strategy The laboratory has set up, in accordance with European positive and negative experiences, the following strategy for developing national standard for length: • the form of the standard and the level in the international hierarchy will remain the same, Figure 3: Gauge block comparators • uncertainty of calibration will be decreased by improving control of environmental conditions and by automating calibration procedures; our reference standards will be calibrated in those European laboratories that can offer the best (lowest) uncertainties; 71 Third International Working Conference “ Total Quality Management - Advanced and Intelligent Approaches “ in two years we plan to reduce measurement uncertainties Such calibrations are cheaper and can be better to the following values: scheduled. Š U = 0,03 μm +0,3⋅10-6⋅ for short gauge blocks, and Š U = 0,05 μm +0,5⋅10-6⋅ for long gauge blocks, 8. CONCLUSIONS • standard basis will be expanding continuously in accordance with the industrial needs, Metrological system in Slovenia is developing very • calibration procedures will be validated fast and is becoming compatible with international periodically by taking part in international interlaboratory system in regulated as well as in non-regulated area. Very comparisons. important part of this development is establishing of This strategy is in perfect compliance with the national standards in order to assure traceability of industrial needs and economical possibilities. industrial measurements. However, the majority of Improvement of the calibration conditions, equipment industry is still not familiar with the approach of this and procedures will be executed through national system. Some companies simply ignore metrology and research programmes. treat it as a “necessary evil”, while the others still live in In the first step (currently in progress) the mechanics the old Yugoslav system where everything was regulated of the long gauge block comparator is being improved by law. Therefore, an efficient education system should and thermal isolation of the gauge blocks during the be established by metrological institutions including the calibration is tested in order to find proper solution. In laboratories maintaining national standards for different the second phase that will start this year, the new Cary physical quantities. Companies should be advised to comparator for short gauge blocks will be automated in introduce in-house calibrations which indeed require high order to eliminate human factor (especially thermal effort, but bring metrological improvement and influence) from the measurement. A concept of a independence. Understanding of metrological principles pneumatic manipulator for loading gauge blocks on the will bring the industry more chance for success on comparator table and for moving table into measurement international market on hand of compatible and positions already exists. The whole system including recognised measurement results. mechanical parts, pneumatics and electronic control will be designed and assembled in the laboratory. REFERENCES

7. TRACEABILITY OF LENGTH /1/ EAL publication EAL-G12. Traceability of STANDARDS IN SLOVENIA Measuring and Test Equipment to National Standards, 1995, /2/ H.J. Warnecke, W.Dutschke. Fertigungsmess-technik, National standard is realised by gauge blocks 0,5 mm Springer-Verlag, ISBN 3-540-11784-9, Berlin 1984 to 1000 mm of quality class 00 res. K. This standard is /3/ Guide to the Expression of Uncertainty in Measurement, ISO Guide, First Edition 1995 being calibrated periodically in one of the European /4/ International Vocabulary of Basic and General Terms national metrology institute (PTB, BNM-LNE, NMi, ...) in Metrology, second edition, Geneva 1993 by interferometry (absolute measurement using laser /5/ EA-4/02. Expression of the Uncertainty of interferometer). Uncertainty of such calibration and the Measurements in Calibration; European Accreditation 1999 facilities of the national laboratory (staff, equipment, /6/ B. Acko. System for assuring traceability of industrial procedures, environmental conditions) allow measurements in Slovenia. Proceedings of the 9th Congress dissemination of the unit of measurement to users International de Métrologie, Barbier, P. (Ed.) pp. 673-676, (industry) with an uncertainty of 0,05 μm +0,5⋅10-6⋅L for ISBN 2-909430-88X, Bordeaux, October 1999, Mouvement Français pour la Qualité; Cedex short gauge blocks (up to 100 mm) res. 0,1 μm +1,0⋅10- 6 /7/ W. Massing. Handbuch Qualitaetsmanagement, Carl ⋅L for long gauge blocks (10 mm to 1000 mm). Hanser Verlag, ISBN, Muenchen Wien 1994 Calibration with such uncertainty can cover calibration needs even for the most precise industry at the moment and in the future. There are neither economical nor technical reasons for involving more precise national standard. Accredited calibration laboratories and industrial companies calibrate their reference (and sometimes also working) standards by comparison with the national standard. Such reference standards are then used for calibration of working standards and measurement equipment. Only few companies use in-house calibrations for assuring traceability of their measurement equipment, what is in fact not very good strategy regarding economical and technical aspect. In-house calibrations should be introduced in all companies managing greater amount of measurement equipment.

72 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

SYNERGY IN APPLICATION OF PRACTICES OF THE INTEGRATED PRODUCTION SYSTEM TOWARDS ACHIEVING MANUFACTURING EXCELLENCE

Keynote paper L Sukarma Agency for the Assesment and Application of Technology, Indonesia

Summary: This paper provides empirical evidence for the existence of synergy in the application of practices of “PIPS” – the Practical Integrated Production System combining TQM, JIT and TPM – or essentially World Class Manufacturing practices. In doing this, the PIPS and company performance – in terms of business and manufacturing performance – are reviewed and hypotheses are developed and verified. It is confirmed that use of Infrastructure practices alone contributes to improved company performance. Moreover, the application of at least one set of unique practices belonging to the PIPS (TQM, JIT or TPM) provides companies a better level of performance than those which only use Infrastructure practices. Ultimately, the use of any combination of TQM, JIT and TPM unique practices or of all the three concurrently, provides companies a superior performance in manufacturing, given adequate use of Infrastructure practices. A subsequent paper will present two illustrations in the application of the PIPS model in interpreting the regression model and accelerating performance improvement through appropriate resource allocation. Keywords: Total Quality Management, Just-In-Time, Total Productive Maintenance, Industrial Improvement Methodology, World Class Manufacturing, Manufacturing Excellence.

1. INTRODUCTION 2. OVERVIEW OF THE PIPS MODEL

The Practical Integrated Production System or PIPS Restated from the previous papers, the PIPS model is (Sukarma and Arndt, 2003a) suggests possible synergy in defined as ‘a combination of a set of principles belonging concurrent application of TQM, JIT and TPM. Empirical to TQM, JIT and TPM, accompanied by a set of evidence confirms the effectiveness of the PIPS model techniques and tools (WCM practices) for achieving (Sukarma and Arndt, 2003b) that plants implementing manufacturing excellence’ (Sukarma, 2000). In this case, TQM, JIT or TPM outperform those which do not manufacturing excellence is ‘a dynamic collection of implement any of them, and that plants implementing all production fundamentals required to generate sufficient three methods concurrently outperform those which capabilities in the framework of satisfying ever-changing implement one or two of the methods. customer preferences in an ever-changing environment.’ Although the above findings are convincing, they are Based on an exhaustive literature survey, the PIPS evaluated based on grouping plants into those adopting model embodies a series of 38 WCM practices, 4 and not adopting a certain improvement method. The indicators of business performance and 15 indicators of problem is that the interpretation of adopting TQM, JIT manufacturing performance (see Figure 1). The 38 WCM and TPM may be different among companies. For practices are grouped into Infrastructure, TQM, JIT and example, one of the respondents in the case study TPM practices. The infrastructure practices consist of interprets ‘adopting JIT’ as ‘call-up order’ or ‘stick to the tools and techniques which an organisation must apply in agreed schedule set together by the customer in supplying order to achieve standard performance in manufacturing a certain amount of products at a certain time’ (Sukarma, and to enable them to support the core approaches, and 2000). This statement is not fully right, since the PIPS hence attain improved company performance. These model defines ‘adopting JIT’ as applying the seven practices include problem solving, employee involvement unique practices of JIT in addition to adequate use of and empowerment, supplier relationships, workplace Infrastructure practices (Sukarma, 2000). management, and other continuous improvement To resolve this problem, performance differentials practices. should be assessed based on grouping plants into The ‘core’ practices for enhancing company applying a set of certain core practices belonging to performance rely on the implementation of TQM, JIT and TQM, JIT and TPM adequately and inadequately, given TPM. The use of TQM and/or JIT and/or TPM is adequate application of Infrastructure practices. This primarily determined by the priority of goals that a firm paper, based on a doctoral thesis at The University of wishes to achieve. If a manufacturer chooses to compete Wollongong (Sukarma, 2000), attempts to demonstrate on delivery and cost, then JIT might be more appropriate. empirical evidence for the existence of synergy in Concurrent implementation of the three methods will application of practices of PIPS, or essentially World undoubtedly result in superior performance. Class Manufacturing practices.

73 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

4 Indicators of Business Performance (Quality, Cost, Delivery and Flexibility) 15 Indicators of Manufacturing Performance

TQM Practices JIT Practices TPM Practices 21. Failure Mode and Effect 27. Set up time Reduction 34. Equipment management & Analysis 28. Focused Factory improvement by teams 22. Design for Manufacturability 29. Group Technology 35. Preventive Maintenance and Quality 30. Pull production system 36. Autonomous Maintenance 23. Taguchi Methods 31. Uniform workload 37. Maintenance prevention 24. Quality Function Deployment 32. Just-in-Time scheduling 38. Maintenance Management 25. Customer survey 33. Kanban System 26. Statistical Process Control

Infrastructure Practices

Other continuous improvement practices 13. Error-proofing (poka-yoke) 14. Quality audits 15. Standardisation of parts, products, and process 16. Cross-functional management 17. Policy deployment (Hoshin Kanri) 18. Visible Improvement Management 19. Benchmarking 20. Value Analysis / Value Engineering

Human Resource Management Problem solving Employee involvement and empowerment 1. B7–basic tools of quality control 4. Employee training 2. N7–advanced tools of quality control 5. Multi-skilled employees 3. PDCA/SDCA 6. Small group improvement activities

Workplace management Supplier management 11. 5S and house-keeping 7. Supplier quality certification 12. Job enlargement/enrichment 8. Reduction of number of suppliers and distances 9. Long term supplier contracts 10. Total supplier evaluation

Figure 1 : Practices of the PIPS Model

Business performance deals with a company’s ability Manufacturing performance concerns the performance to satisfy customers’ needs, hence, customers mostly which companies can measure according to specified perceive their measurement. Juran’s definition of quality standards. It is a collection of yardsticks or key as ‘fitness for use’ (Juran, 1988) can be viewed as a performance indicators in terms of which a company can business performance, since its assessment is mostly assess its ability to satisfy the customer by measuring the based on the customers’ point of view. The performances effectiveness of its resources. This is what a company of Quality, Cost, Delivery and Flexibility are identified should attempt to measure and monitor properly. Juran’s as business performance. definition of quality as ‘conformance to specifications’

74 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

(Juran, 1988) can be regarded as a manufacturing performances. Thus, the first hypothesis can be stated as performance. This article considers the following 15 follows: performances as manufacturing performance: H1: The extent of use of Infrastructure practices 1. In process defects or rework (problem solving, employee involvement and 2. Returns of already-delivered products empowerment, supplier relationships, workplace 3. Manufacturing costs management and other continuous improvement 4. Maintenance costs practices) is positively related to the level of 5. Inventory turnover performance. 6. On-time delivery The crucial role of Infrastructure practices has been 7. Lead time ascertained theoretically and verified empirically in 8. Cycle time several articles but in different situations. The theoretical 9. Space efficiency confirmation can be seen in Sugimori et al. (1977) and 10. Equipment availability Harrison (1994). The empirical evidence can be observed 11. Equipment performance efficiency in Flynn et al. (1995a) for TQM; and Sakakibara et al. 12. Labour productivity (1997) and Nakamura et al. (1999) for JIT. What is new 13. Employee morale and motivation in this paper is the incorporation of TPM as one of the 14. Accident frequency core approaches, besides TQM and JIT. 15. Capital investment efficiency In this context, synergy in performance is achieved when the use of certain core practices leads not only to the improvement in its core performance but also the improvement of other core performance. For example, 3. DEVELOPMENT OF HYPOTHESES Flynn et al. (1995b) found empirical evidence that simultaneous use of JIT and TQM practices might lead The purpose of this paper is to examine the existence not only to the improvement of JIT and quality of synergy in the use of practices of PIPS leading to performance respectively but also to improved additional additional improvement in company performance. Figure performance of quality caused by the JIT practices and to 1 indicates that Infrastructure practices are common to all improved additional performance of JIT caused by the the three methods and their use is aimed primarily at TQM practices. Again, this improvement may be realised supporting the application of the core or unique practices when Infrastructure practices have already been applied. of TQM, JIT and TPM. While the use of Infrastructure By promoting TPM as one of the core approaches, the practices in general affects indirectly to almost all existence of synergy has to incorporate all the three manufacturing performances, and hence to all business methods concurrently. Adapted from Sukarma (2000), performances, the use of core practices aims directly at Table 1 presents a logical explanation of the synergy. improving some specific or core manufacturing

Table 1: Logical Explanations for the Existence of Synergy

Core Company Performance Practices of “PIPS” – Quality JIT TPM Practical Integrated (in process defects, returns (inventory turnover, on-time (equipment availability, Production System of already-delivered delivery, production lead time, equipment performance products) cycle time) efficiency) • Developing quality at • Improving process and • Forcing to improve TQM Unique Practices source assuring quality leading to equipment conditions in • Focusing on customers reduced cycle time, lead order to assure quality time, and inventory parts and finished products • Process improvement • Forcing to improve • Reducing cycle time • Forcing to improve JIT Unique Practices processes due to reduced • Reducing production lead equipment conditions due inventory, production lead time to reduced inventory, time, and cycle time production lead time, and • Reducing inventory cycle time • Reducing equipment • Reducing equipment failure,• Improving equipment TPM Unique Practices related defects leading to equipment set up and availability improved quality parts and adjustments, maintaining • Improving equipment finished products equipment speed leading to efficiency performance reduced cycle time, lead • Improving equipment time and inventory quality performance

75 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

• Improving employees’ ability in problem solving • Involving and empowering employees in decision making Infrastructure Practices • Developing partnerships with suppliers • Improving employees’ ability in workplace management • Others: improving company capability in fail-safing, quality audits, standardisation, cross-functional management, policy deployment, benchmarking, and value analysis and value engineering

To evaluate the existence of synergy in application of performance was also measured using ordinal scales 1) PIPS practices leading to improved company (‘1’ for ‘among the worst’ | ‘2’ for ‘below average’ | ‘3’ performance, two other hypotheses are developed: for ‘average’ | ‘4’ for ‘above average’ | ‘5’ for ‘among H2: The application of at least one set of unique the best’). Regarding the ranking, however, it is practices: TQM (product design, customer focus, process cautioned that the respondent whose answer is ‘4’ cannot management), JIT (set up time reduction, focused factory, be interpreted as having performance twice as much as group technology, pull production system, uniform one whose answer is ‘2’. workload, JIT scheduling, Kanban), or TPM (equipment Since there are only 85 returned questionnaires, it is management and improvement by teams, preventive impossible to relate 38 practices of PIPS with 15 maintenance, autonomous maintenance, maintenance indicators of manufacturing performances and with 4 prevention, maintenance management system), provides indicators of business performances. Therefore, this paper companies a better level of performance than those which examines the relationship between the use of practices of only use Infrastructure practices, given adequate PIPS for each category (Infrastructure, TQM, JIT, and application of Infrastructure practices. TPM) and company performance, which are represented H3: The use of any combination of TQM, JIT and by the average score of business and manufacturing TPM unique practices or of all the three concurrently performances. Using the above measurement, the average provides companies a better level of performance than score of the use of practices as well as business and those which only use one of them, given adequate use of manufacturing performances ranges from 1 to 5. Infrastructure practices.

5. SYNERGY IN APPLICATION OF WCM 4. MEASUREMENT OF THE USE OF PRACTICES ON IMPROVED PRACTICES AND COMPANY PERFORMANCE PERFORMANCE To examine the first hypothesis, it is useful to observe In order to evaluate the above hypotheses, a mail plots of the variables under investigation (see Figures 2 survey on a national basis was designed using methods and 3). developed by Dillman (1978) and conducted by distributing a questionnaire to manufacturing companies in Australia. The questionnaire was designed to gather four kinds of information: (1) general information about the plant; (2) the state of the implementation of improvement programs (TQM, JIT and TPM); (3) application of techniques and their benefits; and (4) performance information systems. The complete description of the questionnaire can be seen in (Sukarma, 2000). The focus of this article is to relate the application of PIPS practices with improved company performance in terms of business and manufacturing performances. In this case, the state of application of each 38 practices was measured in ordinal scales 1) according to its applicability (|‘1’ for ‘never’ | ‘2’ for ‘sometimes’ | ‘3’ for

‘moderately’ | ‘4’ for’ most of the time’ | ‘5’ for 1) ‘always’). Likewise, business and manufacturing Ordinal scales are the next higher level of measurement precision, since variables can be ordered or ranked with ordinal scales in relation to the amount of attributes possessed (Hair et al., 1995).

76 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

3.5

3.0

2.5

ance 2.0

1.5

1.0

verage business perform .5 A 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Average score of infrastructure prac tices

Figure 2 : Business Performance vs. Score of Infrastructure Practices

4.0

3.5

3.0

2.5

2.0

1.5

verage manufacturing performance 1.0 A 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Average score of infrastructure prac tices

Figure 3: Manufacturing Performance vs. Score of Infrastructure Practices

It is apparent from these figures that both business x. In other words, efforts to extend a unit application of and manufacturing performances are associated positively Infrastructure practices may increase manufacturing with the (average) score of Infrastructure practices. performance by 0.436. These two relationships can be analysed further using These models confirm the first hypothesis that the linear regression models. The complete data analysis for level of application of Infrastructure practices is related this hypothesis can be seen in Sukarma (2000). positively to the level of performance. Predictably, The first model shows that business performance (y) Infrastructure practices support the effectiveness of core is correlated positively with score of Infrastructure (TQM, JIT and TPM) practices, and together they practices (x) with a coefficient of 0.581. Analysis of constitute the basis for attaining excellent performance. variance indicates that the regression model is very Investigation of the second hypothesis requires significant. Hence, most of the variation in y can be dividing plants into those which apply Infrastructure explained by the regression equation: y = 0.209 + practices adequately and those which do not, according to 0.463 x. In other words, efforts to extend a unit a certain criterion: application score of Infrastructure application of Infrastructure practices may increase practices is 3.00 or more. 56 plants or 66% of the business performance by 0.463. respondents use Infrastructure practices adequately and The second model reveals that manufacturing 29 plants or 34% not adequately. The 56 plants are performance (y) is also correlated positively with score of grouped further into those which use TQM, JIT and TPM Infrastructure practices (x) with a coefficient of 0.505. practices adequately and not adequately according to the Analysis of variance confirms the significance of the same criterion (application score is 3.00 or more). The regression model, i.e. most of the variation of y can be result is presented in Table 2. explained by the regression equation: y = 1.059 + 0.436

77 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Table 2: The Grouping of Plants into Using TQM/JIT/TPM performance, plants in the second group (using Practices Adequately and Not Adequately Given Adequate Use of Infrastructure and two core practices adequately) perform Infrastructure Practices slightly better than those in the first group (using Not Adequatel Infrastructure and one core practice adequately). For

adequately y manufacturing performance, on the other hand, plants in Groups of Frequen Frequen the first group perform slightly better than those in the % % Total % Practices cy cy second group. TQM practices 17 20.0 39 45.9 56 100 JIT practices 19 22.4 37 43.5 56 100 Table 4: Performances of Groups Using One, Two and Three TPM practices 17 20.0 39 45.9 56 100 Core Practices Adequately Given Adequate Use of Infrastructure Practices Note: A plant is called to use a group of practices Grouping of plants into Group Avera Averag adequate when its application score of practices on that using 1, 2, and 3 core Statistics ge e group is 3.00 or more. practice(s) adequately business manufactur Then using this classification, the t-test for equality of given adequate use of performan ing means can be conducted to investigate the existence of Infrastructure ce performanc difference in performance between these pairs of groups. e Table 3 summarises the results. The complete data Infrastructure + one N 13 13 analysis can be seen in Sukarma (2000). core practice Minim 1.25 1.93 um Table 3: Difference in Performance between Plants Using Maxim 2.25 2.79 TQM/JIT/TPM Practices Adequately and Not Adequately Given um Adequate Use of Infrastructure Practices Mean 1.62 2.31 Std. 0.28 0.26 Significance Deviation Difference in Performance Equal Equal Infrastructure + two N 18 18 varianc varianc core practices Minim 1.00 1.07 es es not um assume assume Maxim 2.50 3.00 d d um Business performance by use Mean 1.58 2.38 of core practices: Std. 0.35 0.50 Use TQM practices .006 .000 Deviation adequately vs. inadequately Infrastructure + three N 22 22 Use JIT practices adequately .057 .026 core practices Minim 1.25 2.00 vs. inadequately um Use TPM practices .031 .005 Maxim 3.00 3.87 adequately vs. inadequately um Manufacturing performance Mean 2.19 2.89 by use of core practices: Std. 0.59 0.59 Use TQM practices .073 .024 Deviation adequately vs. inadequately Use JIT practices adequately .053 .033 To verify difference in performance among these vs. inadequately Use TPM practices .008 .001 three groups, analysis of variance is used. The result is adequately vs. inadequately presented in Table 5.

Detailed analyses (not shown) show that while plants Table 5: ANOVA: Difference in Performance Among Groups Using One, Two and Three Core Practices Adequately Given applying ‘core’ practices inadequately have moderate Adequate Use of Infrastructure Practices performance (larger than 1.50), those which apply core Mean practices adequately have even higher performance (close Sum of df Squar F Sig. to 2.00). Table 3 confirms that performance of plants Squares which apply core practices adequately are significantly e Average better than those which do not. In other words, the Between 36.67 11.10 business 73.358 2 .000 existence of synergy in use of combined Infrastructure Groups 9 9 performance and one of unique practices leading to improved Within 165.094 50 3.302 performance is verified. Groups Again, examination of the third hypothesis requires Total 238.453 52 grouping plants into those which apply one, two and three core practices adequately among those which use Average Between Infrastructure practices adequately. Table 4 presents this manufacturing 3.785 2 1.893 7.631.001 Groups grouping and the performance of each group. performance Within Table 4 shows that business and manufacturing 12.402 50 .248 performances of plants in the third group (using Groups Infrastructure and three core practices adequately) are Total 16.187 52 much higher than the other two groups. For business 78 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

The result shows that there is difference in performance. Predictably, Infrastructure practices support performance among the three groups, but ANOVA can the effectiveness of core (TQM, JIT and TPM) practices, not determine which groups are different. Observing and together they constitute the basis for attaining Table 4, it is useful to compare performance of plants in superior performance in manufacturing. The latter the third group and the combined first and second groups. confirms that performance of plants which apply one core The results are presented in Table 6 for group statistics practices adequately are significantly better than those and Table 7 for t-test for equality of means. which do not. In other words, the existence of synergy in Table 6 shows that average manufacturing and use of combined Infrastructure and one of unique business performances of plants using Infrastructure and practices leading to improved performance is verified. three core practices adequately are much higher than Another new finding shows that average those of plants using Infrastructure and one or two core manufacturing and business performances of plants using practices adequately. Table 7 confirms that performances Infrastructure and three core practices adequately are of these two groups are significantly different. Thus, it much higher than those of plants using Infrastructure and can be concluded that application of all the three core one or two core practices adequately. Thus, it can be practices concurrently and adequately enables companies concluded that application of all the three core practices to achieve a better level of performance than those which concurrently and adequately enables companies to only use one or two of them, given adequate use of achieve a better level of performance than those which Infrastructure practices. only use one or two of them, given adequate use of Infrastructure practices. Table 6: Performances of Groups Using One or Two and Three The above findings seem promising to be Core Practices Adequately Given Adequate Use of implemented in a reality. In order to assist manufacturing Infrastructure Practices companies to utilise these findings, the subsequent paper Grouping of plants into Group Averag Averag will present two illustrations in the application of the PIPS using 1 or 2 and 3 core Statistics e business e model: (1) interpreting the regression model; and (2) practice(s) adequately performan manufactu accelerating performance improvement through given adequate use of ce ring appropriate resource allocation. Both are beneficial in Infrastructure performan attaining improved performance in manufacturing. ce Infrastructure + one or N 31 31 two core practices Mean 1.60 2.35 Std. 0.32 0.41 Deviation REFERENCES

Infrastructure + three N 22 22 /1/ Dillman, D. A. (1978), “Mail and telephone surveys: the core practices Mean 2.19 2.89 total design method”, A Wiley-Interscience Publication, John Std. 0.59 0.59 Wiley & Sons, New York. Deviation /2/ Flynn, B. B., Sakakibara, S. and Schroeder, R. G. (1995b) “Relationship between JIT and TQM: practices and performance”, Academy of Management Journal, Vol. 38 No. Table 7: T-test for Equality of Means between Plants Using 5, pp. 1325-1360. One or Two and Three Core Practices Adequately Given /3/ Flynn, B. B., Schroeder, R. G. and Sakakibara, S. Adequate Use of Infrastructure Practices (1995a) “The impact of quality management practices on performance and competitive advantage”, Decision Sciences, Assumption t df Signifi Vol. 26 No. 5, Sep-Oct, pp. 659-691. cance /4/ Hair, J. F., Anderson, R. E., Tatham, R. L. and Black, W. Average Equal - 51 .000 C. (1995), “Multivariate data analysis with readings”, Fourth business variances 4.755 Edition, Prentice Hall, Englewood Cliffs, New Jersey. performance assumed /5/ Harrison, A. (1994) “Just-in-time manufacturing”, in Equal - 29. .000 Storey, J. (ed.) (1994) New Wave Manufacturing Strategies, variances not4.328 891 Organisational and Human Resource Management Dimensions: assumed An Introduction, Paul Chapman Publishing, Ltd., London, pp. Average Equal - 51 .000 175-203. manufacturing variances 3.923 /6/ Juran, J. M. (1988) “Quality control handbook”, performance assumed McGraw-Hill Publishing Co., New York. Equal - 34. .001 /7/ Nakamura, M., S. Sakakibara, and R. G. Schroeder variances not3.690 853 (1999) “Just-in-Time and other manufacturing practices: assumed implications for U.S. manufacturing performance”, in J. K. Liker, W. M. Fruin, and P. S. Adler (eds.) (1999) Remade in America: Transplanting and Transforming Japanese Management Systems, Oxford University Press, New York. 6. CONCLUSION /8/ Sakakibara, S., B. B. Flynn, R. G. Schroeder, and W. T. Morris (1997) “The impact of just-in-time manufacturing and its infrastructure on manufacturing performance”, Management An examination of the three hypotheses confirms Science, Vol. 43 No. 9, pp. 1246-57. previous researches and obtains new findings. The /9/ Sugimori, Y.K., Kusunoki, F.C. and Uchikawa, S. (1977) former reveals that the level of application of “Toyota production system and Kanban system – materialization Infrastructure practices is related positively to the level of

79 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 of just-in-time and respect-for-human system”, International Journal of Production Research, Vol 15 No 6, pp. 553-564. /10/ Sukarma, L. (2000), A Model for Guiding and Measuring Company Performance Towards Achieving Manufacturing Excellence, PhD Thesis, The University of Wollongong. /11/ Sukarma, L. and Arndt, G. (2003a) “PIPS - the Practical Integration of TQM, JIT and TPM: a Coordinated Approach for Achieving Manufacturing Excellence”, Second International Working Conference, Total Quality Management – Advanced and Intelligent Approaches, Serbia and Montenegro. /12/ Sukarma, L. and Arndt, G. (2003b) “Empirical Evidence for the Effectiveness of the PIPS Model”, Second International Working Conference, Total Quality Management – Advanced and Intelligent Approaches, Serbia and Montenegro.

80 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

CONTINUOUS IMPROVEMENT – A PREREQUISITE FOR SUCCESS Keynote paper R. Pendić1, V. Majstorović2, Z. Pendić3 1 Eurosystems Group – Matheos Invest Group, Belgrade, S&M; 2 Mechanical Engineering Faculty, Belgrade University, S&M; 3 Institute of Nuclear Sciences “VINCA”, Belgrade, S&M

Summary: Although the use of continuous improvement (CI) is widespread today, many organizations are faced with problems associated with both the implementation and long-term sustainability of their CI programs. The aim of this article is to identify factors – enablers – that are important for CI success and longevity. A framework for CI implementation has been adopted, which serves as guideline for practicing managers. At the end, six broad steps are discussed, which, if followed, should create a proper environment for the implementation of CI programs and their success. Keywords: Continous improvement, TQM, Teamwork

1 INTRODUCTION benefits of those small incremental improvements are accomplished over a long period of time frame. The rapidly increasing global competition that many According to Rapp and Eklund (2002), the decline of CI industry sectors world-wide have been facing over the programs in western industry might be why the effects of last 10-15 years, associated with rapid technological CI in organizations have not been as positive as was first development and product variety proliferation, has led to anticipated. the creation of new scenarios according to which The philosophy of CI is to continuously improve, for industries, in order to remain competitive, must instance, organizational processes, services or products. continuously implement best practice management However, as Eklund (1998) argued the benefits from CI principles, strategies and technologies. In that sense, are to give employees an opportunity to participate in many theoretical studies have been published decision-making, express their innovativeness, and emphasizing the importance of strategic management of influence their work environment. various functions and the management of quality in order The purpose of this article is to identify factors that to gain competitive advantage. However, even though are important for the successful implementation of CI some encouraging results have been made in the last few programs and their sustainability. Furthermore, the paper years (Bessant and Caffyn, 1997; Bessant et al., 1994), describes enablers of CI programs, and emphasizes on the Savolainen (1999) stressed out that the research on necessary steps that need to be taken into account in order Continuous Improvement (CI) has not been able to to create a proper organizational environment for the CI produce a sound theoretical basis for this practically- implementation. oriented phenomenon. Although CI is mainly associated with the Total Quality Management (TQM), the practice has become part of the normal functioning organizations (Glover, 2 FRAMEWORK FOR CONTINUOUS 2000). As Rapp and Eklund (2002) argued, numerous IMPROVEMENT forms of organizing and managing CI have been in use, such as quality circles, task forces, suggestion schemes an Companies that are pursuing world-class status may problem solving activities. Even though the practice of CI take different paths that, in turn, require different has received increased attention and interest from the principles. A conceptual is proposed which attempts to western organizations, the results show that they have a facilitate the process of systematically translating harder time succeeding in sustaining the CI programs strategic objectives into improvement programs. This than organizations in Japan, indicating the organizational framework for CI includes three broad steps: strategic culture as the major burden for achieving better results. direction definition; improvement planning; and According to many studies (see Carpinetti and Martins implementation and progress review (Carpinetti and 2001; Brennan, 1991; Griffin, 1988), the implementation Martins, 2001). and the first period of the CI programs are claimed to be In the first step, strategic direction, top managers are successful and have a good results in the organization, it concerned with taking decisions in order to align the has proven difficult to keep up the momentum in the long specific business function (e.g. manufacturing function) run. and related functions with the business strategy envisaged The effects of small incremental improvements (e.g. to obtain long term results. According to Carpinetti and manufacturing processes, productivity, etc.) of CI might Martins (2001), the outputs of this step are competitive not reach full potential until a decline is taking place. priorities, business policies and broad business areas Referring to the experience of CI in Japan, on the candidates for improvements. example of Toyota Production System we can see that the

81 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

The next step in the model, improvement planning, is A performance measurement system can be deployed a process concerned with identifying and prioritizing to establish a system of performance indicators and improvement actions that will most contribute to the methods of driving improvements in performance strategic objectives. Although, improvements can take a efficiency and effectiveness of processes. Proposed and more radical form – process reengineering, this step takes developed by Kaplan and Norton (1996), the balanced the approach of systematically and continuously scorecard as a performance measurement system is deploying strategic improvements. nowadays viewed as a central element of a four-step The third step is concerned with implementing and strategic management system. The balanced scorecard reviewing progress of improvement action. This process supplements measures of financial results with measures is critical since it involves planning, allocation of on three additional perspectives: customers; internal resources, monitoring progress, feedback and if necessary business processes; and learning and growth. These four preventive/corrective actions. perspectives translate a company’s objectives into a set of Another meaningful CI framework has been performance measures. As Kaplan and Norton (1996) suggested by Varghese (2004). This framework for argued, the balanced scorecard is a management system launching a system of improvement processes in an that can motivate critical improvements in such important organization is designed to avoid pitfalls of the process areas as product/service, process, customer, and market improvement investments that yield meaningful results. development. The framework is comprised of four phases (see Table 1.) After establish the critical processes for (Varghese, 2004): improvements, specific improvement actions must be defined, the phase in which benchmarking can be very 1.1 Identify Value Creation useful. Since benchmarking is defined as a process of System continuously measuring and comparing an organization’s 1.2 Inventory enterprise business processes against business leaders (not PLAN 1.3 Determine Process necessarily form the same industry) anywhere in the DATA Collection relevance world, an organization can gain information that will help 1.4 Determine performance it to take actions to improve its performance (Lema and issues Price, 1995). It is important to note that benchmarking is 1.5 Grade by maturity only effective if it is focused on strategic related issues. 2.1 Determine priority DO processes Targeting process 2.2 Establish and deploy process owners 3 TQM AND CONTINUOUS 3.1 Ongoing management IMPROVEMENT CHECK oversight Ongoing oversights 3.2 Align structures and systems Total quality management (TQM) is one of the great management innovations of recent times (Miller, 1996). 4.1 Monitor Process performance As stated earlier one of the most critical concepts of TQM is CI, furthermore to focus on work processes, 4.2 Determine improvement ACT needs analysis of variability, and management by fact (Rapp Manage and improve and Eklund, 2002). Other important principles of TQM 4.3 Launch and manage process are management’s commitment, a focus on customer and interventions the employee, and everybody’s participation (Dahlgaard 4.4 Monitor Process et al., 1998). When applied in the workplace, CI can be performance defined as ongoing improvement involving everyone in the organization – top managers, managers, and workers. Table 1: Continuous Improvement Methodology Furthermore, from the organizational design perspective Phase 1, data collection, collects the information and CI could be defined as “…a purposeful and explicit set of data needed for process targeting, while phase 2, principles, mechanisms, and activities within an targeting processes, uses the information and data from organization adopted to generate ongoing, systematic and phase 1 to make and deploy the targeting decisions. Phase cumulative improvement in deliverables, operating 3 and 4 comprise the activities needed to establish a procedures and systems. CI contributes positively to the system of ongoing or continuous process management organization’s target achievement” (Lillrank et al., 2001). and improvement. Of the many tools and methodologies advocated for 3.1 Quality Circles improvement planning, benchmarking and performance Referring to the earlier discussion regarding tools measurement deserve special attention due to their used to promote CI, one of the most popular are QCs due potential benefit for integrated improvement actions and to the potential benefits and success that an organization efforts towards specific objectives. Another important may encounter. tool, that is widely accepted and used for such activities, The original Japanese organizational vehicles for is quality circles (QC), which will be discussed further promoting CI were Quality Control Circles (QCCs), on. small groups working on improvement projects, and a system called policy deployment, with circles aimed at

82 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 breaking down strategic objectives to operational project 4 ENABLERS OF CI PROGRAMS goals. The QCC is a group of employees from the same work unit who gather regularly to study, discuss and The research shows that the process of moving solve work-related problems using a set of statistical tools towards full CI capability involves acquiring and and following a general problem-solving format (Lillrank embedding key behaviors and is essentially a learning et al., 2001). Solutions are written into suggestions, process (Savolainen, 1998). For instance, the transition presented to management and eventually implemented to from level 1 (No CI activity) – essentially an ad hoc become a part of standard operating procedures. approach – to level 2 (Trying out the new ideas) involves The western version of QCC is Quality Circles (QCs) setting up a systematic approach which provides and this is characterized as a parallel structure separated structures which enable basic problem finding and from the regular activities within the organization. solving behavior to take root (Bessant and Francis, 1999). Groups are composed of volunteers from a work area that For CI to be at all feasible as a potential solution, meet for a couple of hours every week or month (Rapp some fundamental conditions must be met. The and Eklund, 2002). organization must have a basic ability to unfreeze, or to In a literature study by Sheffield et al. (1993) of 448 challenge status quo, (Levin, 1951) its existing principles QC-related articles, the following list of factors fostering and practices, and envision better ways. According to and hindering QCs was created: top management Lillrank et al. (2001), there must also exist some slack commitment, training of participants for involvement in resources that can be invested CI, since an organization QCs, middle management commitment, voluntary spending all of its available resources on a mere survival participation, clear QC objectives, adherence to the QC is not a good place to start CI programs. process, feedback to QCs, and rewards offered for QC After assuming that CI is feasible and desirable, there involvement. are number of requirements that must be met by the Rapp and Eklund (2002) suggest that in order for the organization. This process does not take place overnight step in introducing QC program to be feasible, an but involves considerable learning and fine tuning of the organization must choose how to train and educate mechanisms used to enable CI behavior and create a employees and, further, to select appropriate facilitators positive organizational environment that would enhance and leaders guiding the QC weekly meetings. Other the effects of the implemented CI programs. important factors for implementation of QC programs First, the motivation of individuals and teams must are: voluntary participation, training for members, exist (Lillrank et al., 2001): Do they want to invest time leaders, facilitators and managers, people building and energy in improvement activities? If CI requires philosophy, problem selection and solution by members effort and produces results, there is a need for the of the circle not management, teamwork, positive and appropriate incentive and reward systems designed to visible manager support, and the monitoring activities compensate for extra time and effort, and reward valuable (Lillrank et al., 2001). suggestions and/or ideas (Lillrank et al., 2001; Bessant However, there are numerous problems that might and Francis, 1999). Incentives and rewards can be in form occur as the result of QCs being a parallel organization, of social recognition, money, career insufficient number of volunteers, inadequate training, opportunities/advancements or whatever is considered inability to learn procedures, lack of funding for valuable in a particular organizational culture. meetings, tension may occur because of how people are Second, individuals or groups involved in CI must be treated in the QCs, not all employees may participate equipped with relevant facilities, skills and equipment: leaving them feel left out, etc. Also, a major problem for Can people perform the required tasks? According to the QCs success is the lack of commitment from middle Rickards (1998) and Kobayashi (1990), CI may require managers, who perceive QCs programs as a threat or as training in problem finding and solving process, CI tools irrelevant and a waste of time and resources (Brennan, and techniques, teamwork, statistical analysis and cont- 1991). Furthermore, middle managers may feel that the accounting, among others, reflecting the nature and the workforce is not able to contribute to the solving of magnitude of the task to be resolved or the scope of the organizational problems or they may distrust them. desired CI program. The resolution of these problems lie in the support for Third, people need to know what they are expected to management that should be shown by participation in QC do. In most working environments there is a large number activities and by showing obvious and sincere interest in of improvement targets (Berger, 1997). Priorities and the workers and the problems on which they focus. goals must be set in a manner not to waste organizational Additionally, managers should not put pressure on QCs resources on a less important areas or projects with for immediate results, since the work must take the insignificant payoffs. The organizational and task necessary time. structure of the activity must be defined, the intended An important issue in a QC program is for directions communicated to the employees, not only by management to give quick, knowledgeable and positive the policy statement but also by implementation decisions feedback about what measures to take with the proposed and rewards in order to encourage such behavior. solution. Another critical issue is to implement the The enablers of CI programs and behavior should proposed solution shortly afterwards (Bessant et al., include other elements, such as the following: 1996). Participants need the feedback and motivation of - setting up relevant vehicles (e.g. quality circles) to seeing their suggestions in action, since if the suggestions enact CI (Lillrank and Kano, 1990; Dale, 1995; Berger, are not implemented, the program and management loses 1997); its credibility.

83 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

- development of an idea management system to groups. The questions that should be raised at this point receive and respond to ideas (Schuring and Luijten, include: What do you want or need most from our 1998). product/service? What do we need to make this happen? What product/service features improvements would truly impress you? The next step is to transform the answers on 5 SIX STEPS FOR THE SUCCESSFUL CI these questions into product/service expectations. IMPLEMENTATION It is crucial to ensure that everyone understands the importance behavior standards and that those standards Referring to the previous discussion, in the section we relate to their jobs. The following steps should be offer six basic steps in order to achieve better results of followed: CI implementation. • take one behavior at a time (in the case of automobile As we indicated at the beginning of the paper, not all industry: assembling speed); attempts of CI implementation have been successful, • focus on a selected behavior for three to four months especially before 1990s. There are several elements that • create clear standards (e.g. car assembly time must be explain the failure of these CI programs/strategies (Scott ten hours), scripts, and protocols et al., 2001): • make sure that everyone follows the guidelines all the • many managers though that employee attitude about time. product/service would improve if product/service standards were drafted and distributed for employees 5.3 Identify and eliminate barriers and obstacles to follow or if frontline staff were sent to training Policies, procedures, processes, rules, norms, and programs; protocols that make a negative effect on the • some organizations did not provide their managers organizational performance and make satisfying with the necessary skills, tools, and support needed to customers difficult all need to be eliminated to improve products/services or processes; and successfully improve product/service. Each employee • general lack of understanding that in order to have the must find simpler ways of using their energy and spirit of product/service, the organization needs a creativity in order to completely eliminate these barriers. system for product/service. However, it is important to note that not everything is However, here we offer six steps for success in an a systems or process problem. As Sethi et al. (2002) and effort to create and sustain a culture of CI. Bond (2004) noted that lack of team spirit and relationships within an organization represent the biggest 5.1 Raise awareness barrier in many western organizations. Lack of teamwork It is important to make sure that everybody in is unhealthy and interferes with the quality of product or organization (e.g. employees, managers, stakeholders) customer service. Although many employees and understand why the system is focused on improving managers in these types of organizations accept the different elements of product/service, how this focus is reality and believe that it can not be changed, the solution related to the strategic plan, and what the system is does exist. We need to have common goals and expecting to achieve specifically. This message must be understand how the work we do fits with everyone else’s communicated in a variety of print materials and face-to- work. Managers must insist on great teamwork and face meeting; however the message must not be eliminate issues and barriers that create competition and communicated only for a short time period and then fragmentation. The implementation of proper incentive forgotten since in such case it will not have any effect and reward system that would value and promote whatsoever. Each staff meeting and/or event should teamwork, should be considered as well. present an opportunity to share and negotiate current/future initiatives the organization is 5.4 Learn and develop skills pursuing/planning and the results that are Clearly, employees at all levels have the opportunities accomplished/expected. In short, CI should become an to learn new skills related to product/service excellence ongoing topic for every meeting and forum. (Kobayashi, 1990; Bessant and Francis, 1999). Communication skills such as listening, interviewing, and 5.2 Set service expectations and standards of providing information and problem-solving skills such as behavior dealing with difficult people and managing conflicts are Everyone in the organization needs to know exactly examples of important skills in a service industry. what to do to achieve the system’s goals, meaning that However, the professional educational system must be managers must set clear product/service expectations and developed and implemented in order to develop these standards of behavior, beginning by defining excellent skills in a ways that will achieve results. The product/service. organizations must switch from the old-style approaches According to Scott et al. (2001), the major reason for to training, and learn how to facilitate new techniques the failure of the earlier CI programs was because no that get learning out of the classroom and into the service standards were established or the standards set workplace. For instance, self-directed learning and peer were not specific and clear. In that sense, managers need coaching can become the foundation for skill building, as to involve staff in identifying internal and external managers discover ways to bring staff together for short, customers and discovering what matters most to these targeted, just-in-time learning sessions. Again, we must

84 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 stress out that these learning should be tied into specific important factors that deserve a special attention by behaviors that are being developed. management.

5.5 Listen to your customers REFERENCES Analysis and measurement of customer satisfaction can be a powerful tool to improve product/service if the /1/ Bessant, J. and Francis, D. (1999) Developing strategic feedback is not focused on punishment. Managers need to continuous improvement capability. International Journal of help staff hear, first hand, how they are doing. According Operations & Production Management, 19(11), pp. 1106-1119. to Scott et al. (2001), for this feedback to be effective, it /2/ Bessant, J. and Caffyn, S. (1997) High-involvement must be tied to specific targets and goals and needs to be innovation through continuous improvement. International frontline driven. Employees must identify whom to talk Journal of Technology Management, 14(1), pp. 7-28. to, discover ways to collect and share the information, /3/ Bessant, J. et al (1993) Continuous improvement in and, most important, determine how to use the feedback British manufacturing. Technovation, 13(4), pp. 241-254. constructively to correct their actions and make some /4/ Brennan, M. (1991) Mismanagement and quality circles: further suggestions/proposals. how middle managers influence direct participation. Employee Relations, 13(5), pp. 22-32. However, according to many academics and /5/ Carpinetti, L. C. R. and Martins, R. A. (2001) Continuous practitioners this is an area of great controversy because improvement strategies and production competitive criteria: some managers think that external customer-satisfaction some findings in Brazilian industries. Total Quality surveys are enough, but this is not correct. Micro Management, 12(3), pp. 281-291. improvement efforts need input that is targeted, specific, /6/ Cole, R. E. (2002) From continuous improvement to and in real time. It is also important to understand what continuous innovation. Total Quality Management, 13(8), pp. the customers’ experiences are with the product/service 1051-1056. so the organization can react to potential product/service /7/ Dahlgaard, J.J. et al (1998) Fundamentals of Total Quality problems before they occur, rather than wait after the Management (Great Britain, Chapman & Hall). problems have taken place. If mistakes are an opportunity /8/ Eklund, J. (1998) Participative problem solving and to grow and learn, then organizations need to view them organizational congruence. In: H. Boer and J. Gieskes (Eds.), as the opportunity to improve their performance. Proceedings of the 2nd International EuroCINet Conference in Continuous Improvement: from Idea to Reality, The Netherlands, 14-15 September 1998. 5.6 Reinforce and support CI /9/ Glover L. (2000) Neither poison nor panacea: shopfloor The real challenge for every organization is keeping responses to TQM. Employee Relations, 22(2), pp. 121- 145. the energy, learning, and improving ongoing activity. /10/ Griffin, R.W. (1988) Consequences of quality However, managers finally understand that each staff management in an industrial setting: a longitudinal assessment. meeting and/or event represents an opportunity to learn, Academy of Management Journal, 31(2), pp. 338- 252. /11/ Kaplan, R. and Norton, D. P. (1996) Using the balanced teach, and reinforce the importance of good product scorecard as a strategic management system. Harvard Business quality or customer service. As stated earlier, challenging Review, 74, pp. 75-85. status quo, eliminating inefficient organizational process, /12/ Lema, N. & Price, A. (1995) Benchmarking – encouraging teamwork, implementing appropriate performance improvement toward competitive advantage, education and training, creating proper motivational Journal of Management Engineering, 11, pp. 28-37. programs, and other, represent the prerequisites for the /13/ Lillrank, P. and Kano, N. (1990) Continuous Improvement successful implementation of CI programs in any (Ann Arbor, MI, The University of Michigan Press). organization. /14/ Lillrank, P. et al (2001) Continuous improvement: exploring alternative organizational designs. Total Quality Management, 12(1), pp. 41-55. 6 CONCLUSION /15/ Miller,W. J. (1996) A working definition for TQM researchers, Journal of Quality Management, 1, pp. 149-159. /16/ Rapp, C. and Eklund, J. (2002) Sustainable development This paper has briefly outlined some characteristics of of improvement activities – the long-term operation of a CI, and the actions needed to be taken for its successful suggestion scheme in a Swedish company. Total Quality implementation. Recent survey by Varghese (2004) Management, 13(7), pp. 945-969. reported that most of the western organizations are now /17/ Savolainen, T. I. (1999) Cycles of continuous actively seeking to implement CI in some form, as a improvement: realizing competitive advantages through quality. strategic toll for gaining competitive advantage. International Journal of Operations & Production Our experience in this area [21] shows that the most Management, 19(11), pp. 1203-1222. critical factors for successful CI implementation and its /18/ Scott, G. et al (2001) Customer satisfaction: six strategies longevity are identified as management commitment and for continuous improvement. Journal of Healthcare effective teamwork. Organizations must understand the Management, 46(2), pp. 82-86. real benefits of CI, and put more time in enabling the /19/ Sheffield, D. T. et al (1993) An industry-specific study of necessary factors for its success. Although financial factors contributing to the maintenance and longevity of quality circles, British Journal of Management, 4, pp. 47- 55. resources are required, they are not the most important /20/ Varghese, C. (2004) Resolving the process paradox: a element. Training and education, motivational programs strategy for launching meaningful business process (e.g. recognition), teamwork and the willingness of improvement. Cost Engineering, 46(11), pp. 13-21. employees to express their creativity and innovativeness /21/ Authors’ experiences in implementation of IMS in several in problem solving activities represent much more organizations in Serbia since 1997 till nowadays

85

Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

ACCELERATING PERFORMANCE IMPROVEMENT THROUGH APPROPRIATE RESOURCE ALLOCATION

L Sukarma Agancy for the Assessment and Aplication of Technology, Indonesia

Summary: In order to assist manufacturing companies to gain benefits from the findings of the author’s previous paper, this paper presents two illustrations in application of the PIPS model: (1) interpretation of the regression models relating the use of WCM practices to improved company performance; and (2) acceleration of performance improvement through appropriate resource allocation. The first allows managers to estimate the amount of improved company performance based on their efforts in using the practices belonging to TQM, JIT and TPM. The second enables to guide managers to select the appropriate practices leading to optimum condition of improved company performance. Both are useful in attaining superior performance in manufacturing. Keywords: Total Quality Management, Just-In-Time, Total Productive Maintenance, Industrial Improvement Methodology, World Class Manufacturing, Manufacturing Excellence.

1. INTRODUCTION Average .118 .137 .159 .079 score of TQM The author’ previous paper in this conference provides practices empirical evidence for the existence of synergy in the Average .068 .233 .051 .433 application of practices of “PIPS” – the Practical score of JIT Integrated Production System combining TQM, JIT and practices TPM – or essentially World Class Manufacturing Average .055 .427 .087 .269 practices. It is concluded that the level of application of score of TPM Infrastructure practices is related positively to the level of practices performance; and that application of any combination of TQM, JIT and TPM unique practices or of all the three (1) Dependent Variable: Average Business concurrently, provides companies a superior performance Performance in manufacturing, given adequate use of Infrastructure (2) Dependent Variable: Average Manufacturing practices (Sukarma, 2005). Performance A reasonable further question is that how do managers From Table 1, the regression equations can be written apply the above findings to enable them to interpret as: regression models, and hence to allocate resources BP = 0.254 + 0.234 INF + 0.118 TQM + 0.068 JIT + appropriately? This paper, based on a doctoral thesis at 0.055 TPM ………………. (1) The University of Wollongong (Sukarma, 2000), provides MP = 1.123 + 0.150 INF + 0.159 TQM + 0.051 JIT + two illustrations in the application of the PIPS model. 0.087 TPM …………….... (2) Where: BP = Average Business Performance MP = Average Manufacturing Performance 2. UNDERSTANDING THE REGRESSION INF = Average score of Infrastructure MODELS practices TQM = Average score of TQM practices JIT = Average score of JIT practices The discussion of the author’s previous paper leads to TPM = Average score of TPM practices the development of two regression models, as follows:

The regression models attempt to develop ‘linear’ Table 1: Coefficients of the Regression Equations of the Two Models relationships between concurrent application of Infrastructure, TQM, JIT and TPM practices and Dependent Variables performance. The models are constructed based on the BP (1) MP (2) “Enter” method, that is, all four independent variables are Independent Coefficients Sig Coeffi Sig. included simultaneously to predict performance. Detailed Variables . cients calculation can be seen in Sukarma (2000). (Constant) .254 .274 1.123 .000 The first model shows that the coefficient Average .234 .047 .150 .261 determination (R-Square) is 0.386 and the significance is score of 0.000. This coefficient indicates that about 39% of infrastructure variation in business performance can be explained by the practices

87 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 model and that the model is useful in the sense that at targets, are very powerful for guiding and measuring least one coefficient in Equation (1) is not equal to zero. company performance towards achieving manufacturing Similarly, the second model shows that the coefficient excellence. determination is 0.319 and the significance is 0.000. This means that about 32% of variation in manufacturing performance can be explained by the model and that the model is useful in the sense that at least one coefficient in 3. RESOURCE ALLOCATION TO Equation (2) is not equal to zero. ACCELERATE PERFORMANCE To interpret Equations (1) and (2), it is necessary to IMPROVEMENT observe the significance of each coefficient in Table 1. In Equation (1), the average score of Infrastructure practices The second application is concerned with resource is the only independent variable having a significance allocation in order to accelerate improvement in level less than 0.05. This means that application of manufacturing performance. How should a company Infrastructure practices is linearly related to business measure its performance in order to accommodate the performance. In other words, in the context of concurrent implementation of WCM practices and to foster application of practices, efforts to extend one unit continuous improvement in products and processes? application of Infrastructure practices may increase Moseng and Bredrup (1993) argue that the average business performance by 0.234. This result does measurement of company performance should not necessarily suggest that TQM, JIT and TPM practices incorporate three dimensions: effectiveness, efficiency, do not have any contribution to business performance. and adaptability. The first two criteria have been Table 1 indicates that coefficients of these independent frequently cited in the literature (e.g. Sink and Tuttle, variables are all positive. But their impact on business 1989). The third indicates the extent to which a company performance does not have to be linear. Infrastructure prepares for future changes. This paper incorporates practices contribute the most to performance. This is these criteria in the measurement of manufacturing followed by TQM, JIT and TPM practices respectively. performance. In Equation (2), the constant and average score of Sink and Tuttle (1989) define effectiveness as an TQM practices are the independent variables having a indicator which assesses company activities pertaining to significance level less than 0.10. This result means that doing the right things, at the right time, with the right plants not applying any WCM practice have a quality. Rolstadäs (1998) interprets effectiveness as the manufacturing performance of 1.123 (‘low performance’); ratio between the actual and the expected output. The and at 10% confidence level, it is believed that application fifteen indicators of manufacturing performances in the of TQM practices is linearly related to manufacturing previous paper (Sukarma, 2005) incorporate the performance. In other words, in the context of concurrent effectiveness criteria under the measures of quality application of practices, efforts to extend one unit management (items 1 and 2), equipment maintenance application of TQM practices may increase average (items 10 and 11), HRM (items 12 and 13), and accident manufacturing performance by 0.159 units. This result frequency (item 14). Measures of company performance does not necessarily suggest that Infrastructure, JIT and are interrelated; in the sense that quality management TPM practices do not have any contribution to performances are not merely affected by TQM practices manufacturing performance. Table 1 indicates that the but also by several other practices (particularly HRM and coefficients of these independent variables are all positive. TPM practices). Hence, a company has to consider using But their impact on manufacturing performance does not a combination of practices simultaneously leading to the have to be linear. TQM practices contribute the largest to expected output (superior performance) in order to allow performance. This is followed by Infrastructure, TPM and it to stay in business or to win orders. JIT practices respectively. While the traditional system is concerned mainly with While the regression models developed in examining the efficiency of each individual work centre, Sink and the first hypothesis relate application of Infrastructure Tuttle (1989) define efficiency as the ratio between practices to performance, the models developed in this resources expected to be consumed and actually section relate concurrent application of practices to consumed. Items 3, 4, 5, 9 and 15 of the manufacturing performance. Therefore, the latter models are more performances represent the efficiency criteria. The first indicative of performance than the former. It can be seen two items (3 and 4) are measures of efficiency in that the ‘coefficient determination’ (R-Square) of the spending resources relating to manufacturing and latter models is larger than those of the former. This equipment maintenance respectively. Companies should indicates that the latter can explain variation in allocate their resources properly in order to gain an performance better than the former. optimal cost of quality. In addition, items 5 and 15 are to As argued in Sukarma (2000), the relationship assess efficiency dealing with inventory and capital between company performance and determinants of investment. Lastly, space efficiency (item 9) is a performance is very complicated. The regression models measure of efficiency relating to the use of space. Again, elaborated in this section can explain only one-third of these measures of performance reflect the result of variation in performance. The other factors contributing applying several practices rather than a single technique. to performance can not be incorporated into the regression Last but not least, a manufacturer has to be sensitive models. Guidelines for implementing the PIPS, which and adaptive (agile) to the changing environment in order have been discussed theoretically and validated to survive and prosper in the marketplace. Hamel and empirically in Sukarma (2000), e.g. setting performance

88 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Prahalad (1994) describe competition for the future as an Furthermore, the practices of PIPS aim at improving arena to create and dominate emerging opportunities, not manufacturing performance via better management of just to benchmark a competitor’s products and processes resources existing in a company, which can be simplified and imitate his methods. Items 6 to 8 of the into: people and others (e.g. equipment, materials, manufacturing performances (on-time delivery, energy, and information). Hence, Equation (3) can be production lead time, and cycle time) depict company written as: abilities to adapt to changing conditions. As stated in ∆MP = α (∆MP1 + ∆MP2 +∆MP3) + β (∆MP1 + ∆MP2 Sukarma and Arndt (2003a), delivery reliability is the +∆MP3) ultimate result of a long series of steps performed with α + β = 1 correctly (Maskell, 1991). In other words, it requires a ……………..………………………………..…………. continuous effort to eliminate waste to reduce cycle time … (4) and to shorten lead time at every stage of production, in where α: relative addition to correct scheduling, quality assurance, and on- contribution to manufacturing time deliveries of parts and components. performance due to improved Moreover, optimal conditions are accomplished management of people; and through appropriate allocation of resources which lead to β: relative contribution to the greatest improvement, ∆MP, vis a vis: effectiveness manufacturing performance due (∆MP1), efficiency (∆MP2), and adaptability (∆MP3), or to improved management of ∆MP = ∆MP1 + ∆MP2 + ∆MP3 others. …………………………………….………….. (3) As seen in the PIPS Framework (Sukarma and Arndt, Table 2 shows the grouping of manufacturing 2003a), HRM plays a critical role in managing performance according to these three indicators. It is also Infrastructure and in supporting the core approaches. argued that plants should be concerned first with Hence, improved management of people contributes to effectiveness, then with efficiency and adaptability or vice improved management of others. In other words, β in versa, depending on the circumstances. Equation (4) contains contribution of people. As explained earlier, after separating the people component Table 2: Manufacturing Performances and Indicators of from β, Equation (4) can be written as: Performance ∆MP = α1 (∆MP1 + ∆MP2 +∆MP3) + β1 (∆MP1 + ∆MP2 Indicators of: +∆MP3) where α and β are contributions of ‘all people’ and Manufacturing Eff Ef Adap 1 1 ‘others’ respectively; or Performance ectiven ficien tability ∆MP = ∆MP + ∆MP ess cy people others …………………………………………….. (5) 1. In process defects ● ‘How can improvement in management of people be or rework accelerated?’ Although a company’s success in 2. Return of ● improving performance is determined by many factors, already-delivered the management of people is the most “tangible” and products most easily influenced (Arndt, 1985). If the contribution 3. Manufacturing ● to improved performance due to improved management costs of people (α ) is split into education (α ), training (α ), 4. Maintenance ● 1 11 12 and others (α13), then the first part of Equation (5) can be costs written as 5. Inventory ● ∆MP people = (α11 + α12 + α13) (∆MP1 + ∆MP2 + ∆MP3) turnover ...………….….……. (6) 6. On-time ● Since formal education and other measures (e.g. delivery employee relations) can only lead to long-term 7. Lead time ● improvement, the best way to improve performance (production) ‘quickly’ is via industrial (e.g. on the job) mass-training. 8. Cycle time ● This is particularly true in the case of improving the 9. Space efficiency ● ‘effectiveness’ indicator of the performance (∆MP1). 10. Equipment ● Thus, Equation (6) can be written as: availability ∆MP people = α12 ∆MP1 + α12 (∆MP2 +∆MP3) + 11. Equipment ● (α11+α13) (∆MP1 +∆MP2 +∆MP3) …..… (7) performance efficiency or 12. Labour ● ∆MP k α12 α12 (α11+α13) productivity people ∆MP1 (∆MP2 (∆MP1 +∆MP2 13. Employee ● +∆MP3) +∆MP3) morale and motivation Improve Improve Improveme Improvement in 14. Accident ● ment in ment in nt in all indicators frequency manufact effective efficiency due to education 15. Capital ● uring ness due and and other people investment efficiency performa to adaptabilit management nce due training y due to

89 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

to training practices, 4 out of 7 JIT practices, and 3 out of 5 TPM improve practices. These changes modify budget allocation of all d practices from ‘normal’ (.20 .20 .20 .40) to ‘accelerated’ manage (.1697 .3211 .1697 .3395). ment of A practice may be considered to be accelerated when people its effectiveness needs participation of all (shopfloor) employees, not just (limited to) a certain group of In other words, ∆MP people can be accelerated when (highly) educated people. For instance, the effectiveness companies can improve their effectiveness by way of of problem solving practices requires each member to expanding industrial mass-training, or when they can apply seven basic or even advanced tools of quality make k greater than 1. control. Consequently, mass-training in these tools will To illustrate this, suppose that a typical company undoubtedly lead to increased ability of problem solving spends 60% of its improvement budget for improving in the workplace, and hence to accelerated improvement management of people and the remaining 40% for in performance. A training model of some of the improving management of others (equipment, materials, accelerated practices can be found in Arndt (1989 a, b, c, energy and information). Suppose also that the former is d). On the other hand, the practices of benchmarking and distributed as: 20% for education, 20% for training, and product design, for example, can be applied by a certain 20% for other aspects of people management. Expanding group of people, or even can be conducted by external industrial training may mean doubling the percentage of consultants. Therefore, mass-training in these practices is budget spent on training, so the new distribution not necessary. becomes: 15% for education, 40% for training, 15% for Furthermore, Equations (5) and (7) can be combined other aspects of people management, and 30% for as: management of others. ∆MP = ∆MP people + ∆MP others . Table 3 recommends changes of budget allocation on = α12 ∆MP1 + α12 (∆MP2 +∆MP3) + several practices of the integrated model from ‘normal’ (α11+α13) (∆MP1 +∆MP2 +∆MP3) (.20 .20 .20 .40) to ‘accelerated’ (.15 .40 .15 .30). The + β1 (∆MP1 +∆MP2 accelerated practices are printed in italics. These include +∆MP3) ……………………....…...... (8) 13 out of 20 Infrastructure practices, 3 out of 6 TQM

Table 3: Changes in Resource Allocation from ‘Normal’ to ‘Accelerated’ Management of people Management Practices of PIPS Education Training Others of others A. Infrastructure practices Problem solving 1. B7–basic tools of quality control 0.15 0.4 0.15 0.3 2. N7–advanced tools of quality control 0.15 0.4 0.15 0.3 3. PDCA/SDCA 0.15 0.4 0.15 0.3 Employee involvement and empowerment 4. Employee training 0.15 0.4 0.15 0.3 5. Multi-skilled employees 0.15 0.4 0.15 0.3 6. Small group improvement activities 0.15 0.4 0.15 0.3 Supplier management 7. Supplier quality certification 0.2 0.2 0.2 0.4 8. Reduction of number of suppliers & distances 0.2 0.2 0.2 0.4 9. Long term supplier contracts 0.2 0.2 0.2 0.4 10. Total supplier evaluation 0.2 0.2 0.2 0.4 Workplace management 11. 5S and house-keeping 0.15 0.4 0.15 0.3 13. Job enlargement/enrichment 0.15 0.4 0.15 0.3 Other continuous improvement practices 13. Error-proofing (poka-yoke) 0.15 0.4 0.15 0.3 14. Quality audits 0.15 0.4 0.15 0.3 15. Standardisation of products and process 0.15 0.4 0.15 0.3 16. Cross-functional management 0.15 0.4 0.15 0.3 17. Policy deployment (Hoshin Kanri) 0.2 0.2 0.2 0.4 18. Visible Improvement Management 0.15 0.4 0.15 0.3 19. Benchmarking 0.2 0.2 0.2 0.4 20. Value Analysis / Value Engineering 0.2 0.2 0.2 0.4

90 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

B. TQM Practices 21. Failure Mode and Effect Analysis 0.2 0.2 0.2 0.4 22. Design for Manufacturability and Quality 0.2 0.2 0.2 0.4 23. Taguchi Methods 0.15 0.4 0.15 0.3 24. Quality Function Deployment 0.15 0.4 0.15 0.3 25. Customer survey 0.2 0.2 0.2 0.4 26. Statistical Process Control 0.15 0.4 0.15 0.3 C. JIT Practices 27. Set up time Reduction 0.15 0.4 0.15 0.3 28. Focused Factory 0.2 0.2 0.2 0.4 29. Group Technology 0.2 0.2 0.2 0.4 30. Pull production system 0.15 0.4 0.15 0.3 31. Uniform workload 0.15 0.4 0.15 0.3 32. Just-in-Time scheduling 0.2 0.2 0.2 0.4 33. Kanban 0.15 0.4 0.15 0.3 D. TPM Practices 34. Equipment mngmnt & imprvmnt by teams 0.15 0.4 0.15 0.3 35. Preventive Maintenance 0.15 0.4 0.15 0.3 36. Autonomous Maintenance 0.15 0.4 0.15 0.3 37. Maintenance prevention 0.2 0.2 0.2 0.4 38. Maintenance Management System 0.2 0.2 0.2 0.4 All practices (average) 0.1697 0.3211 0.1697 0.3395 Note: ‘Normal’ – (.20 .20 .20 .40); ‘Accelerated’ ( .1697 .3211 .1697 .3395)

As explained earlier, the first part of Equation (8) is Normal 0.1000 0.1000 0.4000 0.4000 the performance that can be achieved quickly. To (Σ=1.00) illustrate this, all items in Equation (8) have to be Accelerated 0.1605 0.1605 0.3395 0.3395 estimated. Table 2 indicates that seven measures of (Σ=1.00) manufacturing performance are indicators of Normal 0.0447 0.0447 0.1788 0.1788 effectiveness, five are indicators of efficiency, and three (Σ=0.447) Accelerated 0.0718 0.0718 0.1517 0.1517 are indicators of adaptability. Hence, it is reasonable to (Σ=0.447) argue that ∆MP1 contributes 50%, ∆MP2 contributes 30%, and ∆MP3 contributes 20% to overall improvement in Table 4 (rows 3 and 4) indicates that changing budget manufacturing performance (∆MP). Assuming that the allocation from ‘normal’ to ‘accelerated’ increases the improved performance is proportionate to budget first part of Equation (8) from 10% to 16% or k = 1.6. allocation, then α11 = .1697, α12 = .3211, α13 = .1697, and This means that companies can improve their β1 = .3395. effectiveness ‘quickly’ by expanding employee training Table 4 presents the calculation of Equation (8) in the practices contributing to effectiveness (see Table according to the two scenarios described earlier: ‘normal’ 3). As stated in Sukarma (2000), companies and ‘accelerated’. implementing TQM, JIT and TPM for up to two years have the highest performance compared to other duration Table 4: Distribution of Manufacturing Performance: ‘Normal’ of implementation. The reason for this, as argued earlier, vs. ‘Accelerated’ is that these recently implementing plants are eager to ∆MP = Α1 + + + β1 provide their employees training on quality-related 2 α12 (α11 + (∆MP1 + techniques to enable them to boost their performance in a ∆MP1 (∆MP2 α13) ∆MP2 + relatively short period of time. + (∆MP1 + ∆MP3) Assuming further that concurrent application of ∆MP ) ∆MP + 3 2 Infrastructure, TQM, JIT and TPM practices is related ∆MP3) Overall Impro Improv Improve Improve linearly to performance, Equation (2) becomes: improvemen vemen ement ment in ment in ∆MP = 0.150 ∆INF + 0.159 ∆TQM + 0.051 ∆JIT + t in t in in all all 0.087 ∆TPM manufacturi effecti efficien indicator indicator This equation implies that one unit increment in the ng veness cy & s due to s due to application of Infrastructure, TQM, JIT and TPM performance due to adaptab educatio improved practices may potentially, although gradually, lead to trainin ility n and manage improved manufacturing performance (∆MP) by 0.447 g due to other ment of unit. By setting ∆MP = 0.447, instead of 1.00 as the training people others previous discussion, the last two rows of Table 4 can be manage ment calculated.

91 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

The explanation for the latter is similar to that for the (limited to) a certain group of (highly) educated people. previous discussion. Instead of gradually improving their In fact, changing budget allocation from ‘normal’ to performance through normal budget allocation, ‘accelerated’ increases manufacturing performance due to companies can speed up improvement in effectiveness by improved management of people from 10% to 16% or k = multiplying training budget on certain practices. In fact, 1.6. This means that companies can improve their improved company capability in ‘effectiveness’ or ‘doing effectiveness ‘quickly’ by expanding industrial mass the right things’ may potentially lead to an increase in training in the practices contributing to effectiveness. other indicators of performance. Table 4 also indicates that expanding the training budget increases not only effectiveness (the first part of Equation (8)), but also efficiency and adaptability REFERENCES indicators of performance (the second part of Equation (8)). Although the latter may be attained gradually, these /1/ Arndt, G. (1985) “Technology transfer and agricultural people-related performances are very useful in robotics”, Annals of the CIRP, Vol. 34 No. 1. maintaining and sustaining continuous improvement. /2/ Arndt, G. (1989a) “Basic improvement concepts and the Finally, increase in first and second parts is compensated ‘basic seven’ tools for TQM”, (JQT - I/PIQS-Basic), CAMIA, for by reduction in the third and fourth parts. This trade- University of Wollongong. off is unavoidable, yet improving effectiveness in the first /3/ Arndt, G. (1989b) “Further ‘Kaizen’ concepts and the place is worthwhile. ‘new seven’ tools for TQM”, (JQT - II / PIQS-Planning), CAMIA, University of Wollongong. /4/ Arndt, G. (1989c) “‘Kaizen’ strategy and implementation using VIM: visible improvement management”, (JQT - III / PIQS-Mgmt), CAMIA, University of Wollongong. 4. CONCLUSION /5/ Arndt, G. (1989d) “An update on ‘Kaizen’ / JQT developments and an introduction to Taguchi methods”, (JQT - Two illustrations in application of the PIPS model are IV / PIQS KJ-Taguchi), CAMIA, University of Wollongong. /6/ Hamel, G. and C. K. Prahalad (1994) “Competing for presented: (1) interpretation of the regression models the future”, Harvard Business School Press, Boston, MA relating the use of WCM practices to improved company /7/ Maskell, B. H. (1991) “Performance measurement for performance; and (2) acceleration of performance world class manufacturing: a model for American companies”, improvement through appropriate resource allocation. Productivity Press Inc., Cambridge, MA The first allows managers to estimate the amount of /8/ Moseng, B. and H. Bredrup (1993) “A methodology for improved company performance based on their efforts in industrial studies of productivity performance”, Production using the practices belonging to TQM, JIT and TPM. The Planning and Control, Vol. 4, No. 3. second enables to guide managers to select the /9/ Rolstadăs, A. (1998) “Enterprise performance appropriate practices leading to optimum condition of measurement”, International Journal of Operations & Production Management, Vol. 18 No. 9/10, pp. 989-999 improved company performance. Both are useful in /10/ Sink, S. and T. Tuttle (1989) “Planning and attaining superior performance in manufacturing. measurement in your organisation of the future”, Industrial From the first illustration, it is found that Engineering and Management Press, Norcross, GA Infrastructure practices contribute the most to business /11/ Sukarma, L. (2000), A Model for Guiding and performance, in which efforts to extend one unit Measuring Company Performance Towards Achieving application of Infrastructure practices may increase Manufacturing Excellence, PhD Thesis, The University of average business performance by 0.234 units. This is Wollongong. followed by TQM, JIT and TPM practices respectively. /12/ Sukarma, L. and Arndt, G. (2003a) “PIPS - the Practical For manufacturing performance, TQM practices Integration of TQM, JIT and TPM: a Coordinated Approach for Achieving Manufacturing Excellence”, Second International contribute the largest to performance, in which efforts to Working Conference, Total Quality Management – Advanced extend one unit application of TQM practices may and Intelligent Approaches, Serbia and Montenegro. increase average manufacturing performance by 0.159 /13/ Sukarma, L. and Arndt, G. (2003b) “Empirical units. This is followed by Infrastructure, TPM and JIT Evidence for the Effectiveness of the PIPS Model”, Second practices respectively. However, the relationship between International Working Conference, Total Quality Management company performance and determinants of performance – Advanced and Intelligent Approaches, Serbia and is very complicated. The regression models elaborated Montenegro. can explain only one-third of variation in performance. /14/ Sukarma, L. (2005) “Synergy in Application of The other factors contributing to performance can not be Practices of the Integrated Production System towards Achieving Manufacturing Excellence”, Third International incorporated into the regression models. Working Conference, Total Quality Management – Advanced From the second illustration, it is found that a practice and Intelligent Approaches, Serbia and Montenegro can be considered to be accelerated when its effectiveness needs participation of all (shopfloor) employees, not just

92 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

APPLICATION OF TAGUCHI MODEL FOR QUALITY PRODUCT IMPROVEMENT

Mech. Eng. B.Sc. Tatjana V. Sibalija1, Prof. Dr. Vidosav D. Majstorovich2 1 STMicroelectronics Malta Ltd., Industry Road, Kirkop, Malta, 2 Mechanical Engineering Faculty, Laboratory for Production Metrology and TQM, Serbia

Summary: Taguchi Design of Experiment model, as a specific quality engineering technique within Taguchi TQM model, will be considered in this paper. The aim of this project is to redesign thermosonic copper wire bonding process, in order to achieve product - microelectronic package quality improvement, simultaneously achieving process cost reduction. After defining the specifications for quality characteristics, selection of microelectronic components, machine, tool and initial process parameter values, process design is preformed using Taguchi orthogonal design technique. Analysis of experimental results provides selection and verification of the adopted process parameters values, leading to product quality characteristics improvement. Keywords: Taguchi Model, Design of Experiment, Thermosonic Copper Wire Bonding, Microelectronic Package.

1 INTRODUCTION copper wire Thermosonic wire bonding is the most widely used assembly technique in the semiconductor industry to interconnect the internal circuitry of the die to the weld external world. This method uses force, power, time, temperature and ultrasonic energy to form the silver lead connections. The main elements in the microelectronic surface package are: die with internal circuitry, and output pins - copper leads covered with silver layer (Figure 1). die

Figure 2: Welds at silver lead surface (magnification 10 x).

copper wire

leads weld

silver layer on lead

copper lead

copper wire - loops Figure 1: Bonded microelectronic package (device). Figure 3: Cross-section of copper lead covered with silver layer with copper wire weld (magnification 40 x). Wire bonding cycle makes one connection between the bond pad and the lead, using wire. A fine copper wire The idea of this project is to redesign copper wire (50 μm diameter) is fed down through the tool, called bonding process to directly weld copper wire to the bare capillary. The ultrasonic transducer converts the electrical copper leads, with no silver layer as a mediator /1/. energy and transmits this resonant energy at the tip of the Considering price of silver and comparing electrical bonding capillary, to form the copper ball. Capillary is and thermical properties of silver and copper, the moving down to the aluminium bond pad placed on the advantages of performing copper wire bonding to the die surface, to form the ball bond. Then, the capillary lifts bare copper leads are /1/: up, forming the looping profile, and comes down to form • elimination of silver from microlelectronic the weld at the lead. This cycle is repeated until the package leads to the process/product cost reduction; microelectronic package is bonded. • since electrical and thermical properties of Due to oxidation of copper at higher temeratures, copper are more favorable for this application than output copper leads are covered with silver layer in order characteristics of silver, performing copper wire bodning to ensure reliable weld between copper wire and ooutput to the bare copper leads improves the quality and the pins (Figure 2. and 3.) /1/. reliability of microelectronic package.

93 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

This paper considers designing of one part of copper - Contact Time = 5 – 20 ms wire bonding process, which refers to forming the weld - Contact Power = 20 - 60 mu between copper wire and bare copper leads, instead of - Contact Force = 50 - 150 gram silver-plated leads. - Base Time = 20 – 40 ms - Base Power = 80 – 120 mu 1.1 Quality Characteristics - Base Force = 260 – 350 gram - Gas Flow = 0.6 – 1 l/min The test, performed to show the integrity of weld - Scrub SPC = “Yes” / “No”, between copper wire and bare copper lead, is known as where ‘mu’ means ‘machine unit’. “Pull test”. As one microelectronic package (device) has Process parameters – preheat and bond-site several loops – wires connecting the die with the output temperatures will not be changed, since that would affect leads, Pull test is performed on all wires in one device, quality of the first bond between copper wire and and then mean value is calculated. Since there is a aluminum pads at the die /1/. specification for lower Pull test value, it is necessary that all Pull test values measured in one device are higher than lower specification limit, therefore another important quality characteristic is minimum pull test 2 DOE – ORTOGONAL DESIGN value measured in one device. Process yield – process efficiency is process quality characteristic which estimates the process robustness. Taguchi's approach, using ortogonal arrays and Process yield is ratio between the number of successufuly fractional factoriel, instead of full factorial, alows setting bonded devices and total number of devices, in one up experiments with a very large number of factors on sample. This ratio takes in account number of few levels, in an easy «user-friendly» maner . connections – wires in one devices. This experiment will be performed with nine factors Quality characteristics, which will be considered as which will be varied on two levels, and three outcomes responses in following experiment, are /1/: (quality characteristics) will be observed. - Pull Test average value (gram), Data input, design of experiment, data processing and - Pull Test minimum value (gram), analysis are generated by using special statistical - Yield (%). software equipped with experimental design module - Pull Test values are measured using special pull tester software «ECHIP”. equipment, where tool-hook is placed beneath the wire; hook then moves vertically, pulling force is increasing 2.1 Design Variables gradually till sensor detects wire break. Using special First step will be the definition of design variables. interface, software processes the data and presents them This experiment will be performed with 9 factors (design at display. variables), where 8 factors belong to continuous type and Lower Specification limit, for wire bonding with 50 factor Scrub SPC belongs to categorical type. μm copper wire diameter, is: LSL = 26 grams. The continuous variables range will be chosen, and used to scale the design. 1.2 Process Parameters After forming the first bond at the die, tool is moving 2.2 Response Variables up, forming loop and going down to the lead surface Measured quality characteristics (response variables) applying Standby Power, to stabilize contact between are: Pull Test average value (gram), Pull Test minimum tool with copper wire and lead contact surface. During value (gram) and process Yield (%). Contact Time, Contact Force and Contact Power are In the experiment, microelectronic package contains applied in order to make the contact between wire and 58 wires; Pull test average value is average value from 58 lead. Then, Base Power and Base Force are applied Pull test readings; Pull test minimum value is minimum during Base Time for final weld formation /2/. measured value from 58 readings, for each trial. If there are impurities present at that contact welding Process Yield is ratio between number of successfuly area, or surface roughness quality is low, it is desirable to bonded wires (welds) over total number, in one sample of perform scrubbing – x-y oscillation at the weld position. 6 devices. In this case, Scrub type SPC is adopted – scrubbing perpendicular to the wire length. When scrubbing is 2.3 Standard Design: Taguchi Orthogonal applied, Standby Power and Contact Power values are set to 0. Array (L12) For thermosonic bonding of copper wire to bare Taguchi orthogonal design is chosen for the design of copper leads, it is necessary to provide reduction this experiment. Since there are 9 control factors, protective gas (N2 95%, H2 5%) in preheat zone above experiment is designed on matrix L12 containing 12 trials, leads, to prevent bare copper lead surface from oxidation plus 5 repetitions (Figure 4). Software will automaticly /1/. Gas Flow rate will be considered as one of the generate plan of experiment - design spreadsheet with 17 process parameters. trials, by randomized layout (Table 1). The array L12 is These nine parameters will be used as control factors equivalent to a 211 full factorial experiment, which would for experiment. Initial working parameters used as require 2048 trials, in classical approach. starting point for process design are /1/: In order to perform the experiment easier, original - Standby Power = 10 - 50 mu plan will be divided in two groups: 94 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

– first group will contain trials with scrubbing: For modified plan, factors codification is shown in the Scrub SPC = “Yes”; where Standby Power and Contact Table 3. Code “–1” presents minimum value for the Power will not be applied; factor; code “+1” presents maximum value for the – trials in second group will be performed without observed factor. scrubbing - Scrub SPC = “No”. There is a balancing property between the columns in Modified plan of experiment is shown in table 2. the Taguchi matrix (Table 3), which guarantees that the conclusion reached about factors will be independent and uncorrelated, which is known as «ortogonality» /3/. Ortogonality means that the scalar multiplying of any two columns, without repetition, is equal to 0. This is 2-level fractional factorial design, and model is called «linear plus», which includes repetitions. After performing all trials in the experiment, measurements and data collecting, response values data should be entered in appropriate table (Figure 5)

Figure 4: Taguchi Orthogonal Design based on matrix L12 plus 5 replicates. .

Table 1: Original plan of experiment based on matrix L12 plus 5 replicates - 17 randomized trials. Control factors trial Standby Contact Contact Contact Base Base Power Base Force Gas Flow Scrub no. Power (mu) Time (ms) Power (mu) Force (mu) Time (ms) (mu) (mu) (l/min) 3 10 5 60 150 40 80 260 0.6 No 7 50 5 60 150 20 80 350 1 Yes 8 50 5 60 50 40 120 350 0.6 Yes 1 10 5 20 50 20 80 260 0.6 Yes 2 10 5 20 50 20 120 350 1 No 5 10 20 60 50 40 120 260 1 Yes 6 10 20 60 150 20 120 350 0.6 No 1 10 5 20 50 20 80 260 0.6 Yes 3 10 5 60 150 40 80 260 0.6 No 4 10 20 20 150 40 80 350 1 Yes 10 50 20 60 50 20 80 260 1 No 9 50 5 20 150 40 120 260 1 No 11 50 20 20 150 20 120 260 0.6 Yes 2 10 5 20 50 20 120 350 1 No 5 10 20 60 50 40 120 260 1 Yes 12 50 20 20 50 40 80 350 0.6 No 4 10 20 20 150 40 80 350 1 Yes

Table 2: Modified plan of experiment Control factors trial Standby Contact Contact Contact Base Time Base Power Base Gas Flow Scrub no. Power (mu) Time (ms) Power Force (mu) (ms) (mu) Force (l/min) 4 0 20 0 150 40 80 350 1 Yes 7 0 5 0 150 20 80 350 1 Yes 1 0 5 0 50 20 80 260 0.6 Yes 5 0 20 0 50 40 120 260 1 Yes 11 0 20 0 150 20 120 260 0.6 Yes 1 0 5 0 50 20 80 260 0.6 Yes 4 0 20 0 150 40 80 350 1 Yes 8 0 5 0 50 40 120 350 0.6 Yes 5 0 20 0 50 40 120 260 1 Yes 12 50 20 20 50 40 80 350 0.6 No 6 10 20 60 150 20 120 350 0.6 No 9 50 5 20 150 40 120 260 1 No

95 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

2 10 5 20 50 20 120 350 1 No 3 10 5 60 150 40 80 260 0.6 No 10 50 20 60 50 20 80 260 1 No 3 10 5 60 150 40 80 260 0.6 No 2 10 5 20 50 20 120 350 1 No

Final ANOVA tables for Pull Test average value, Pull Control Test minimum value and process Yield (%), respectively, Factors are shown in tables 4, 5 and 6. Calculation of averages values of responses contains

Standby Power Contact Time Contact Power Contact Force Base Time Base Power Base Force Gas Flow Scrub totals for each level of factors in the design. Totals are 1 2 3 4 5 6 7 8 9 calculated by looking at the modified plan of experiment columns (Table 2.), and by adding the outcomes for each level of 1 / -1 / -1 -1 -1 -1 -1 -1 all the trials that have been performed with that factor 4 / +1 / +1 +1 -1 +1 +1 -1 level, shown at figure 5. Mean values are calculated by 5 / +1 / -1 +1 +1 -1 +1 -1 dividing the total with the number of trials for that level, 7 / -1 / +1 -1 -1 +1 +1 -1 for particular factor. 8 / -1 / -1 +1 +1 +1 -1 -1 Mean squares are calculated using following formulas 11 / +1 / +1 -1 +1 -1 -1 -1 /3/: 12 +1 +1 -1 -1 +1 -1 +1 -1 +1 2 2 2 * Total: STi = I1 +... + In – ((ΣI) / n) 2 -1 -1 -1 -1 -1 +1 +1 +1 +1 2 2 * Factor: SFI = (ΣIF-1) / n F-1 + (ΣIF+1) / n F+1 – (ΣIF-1+ 3 -1 -1 +1 +1 +1 -1 -1 -1 +1 2 6 -1 +1 +1 +1 -1 +1 +1 -1 +1 ΣIF+1) / n 9 +1 -1 -1 +1 +1 +1 -1 +1 +1 * Error : Serror = STi – Σ SFI , where: 10 +1 +1 +1 -1 -1 -1 -1 +1 +1 - I - response; - n=17 - number of trials; Table 3: Modified Taguchi matrix L12. - F - control factor;

- ΣIF-1 / ΣIF+1 – sum of responses where factor F has minimum / maximum value; - n F-1 / n F+1 – numer of trials where factor F has minimum / maximum value;

- Σ SFI – sum of mean squares of all factors, for response I. First coloumn of ANOVA tables (tables 4, 5 and 6), shows mean squers for each factor, for “within experiment” error and for replicate error /4/. Taking in account total mean squares and individual mean square values, one can calculate factor variation share in total response variation, for each factor (P-values). In the second column of ANOVA tables (tables 4, 5 and 6), one can see degrees of freedom for each factor and for replicate. When repetitions are included, there is another kind of variation, namely “within experiment” error. The ERROR line describes all degrees of freedom Figure 5: The Response Values Spreadsheet. not expressed on the other lines. Some of these degrees may replicate degrees of freedom, what the REPLICATE 2.4 Analysis ERROR line breaks out /4/. Residuals values can be found by subtraction of the corresponding sums of The analysis consists of fitting a model to the data. squares. The result is a set of coefficient estimates, which Degree of freedom for error equals 5 (5 repetitions), summarizes the relationship between the design and degree of freedom for each factor is 1 due to two-level response. Software will analyze this design and create factor variation (2 levels–1=1). Degree of freedom for ANOVA table, which is a custom method of analysis error contents all degrees of freedom not comprehended ANOVA with degrees of freedom for factors and replicate error. In With the increased number of factors, it is necessary this experiment, all 9 factors have 1 degree of freedom, to remove those that are not significant for certain replicate error has 5 degrees of freedom. In total, there response, so that better estimate of the relative are 16 degrees of freedom (17 trials–1=16). The other 2 importance of each of the significant factors can be degress of freedom, not assigned by factors and replicate obtained. This process is known as “pooling”. error, present degree of freedom for ”within experiment” Insignificant factors are pooled, and this implies that the error. variation due to these factors was no greater than that due to unassignable error. Remaining factors are forming final ANOVA table.

96 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Response: Yield Response: Pull test average value Mean Degrees of P Control Factors Squares Freedom Degrees Mean Squares of P Control Factors 75.8315 1 0.1073 Standby Power Freedom 58.4183 1 0.1491 Contact Time 27.0257 1 0.0004 Standby Power 157.848 1 0.0323 Contact Power 27.7321 1 0.0004 Contact Time 457.034 1 0.0027 Contact Force 0.047995 1 0.7970 Contact Power 468,004 1 0.0025 Base Time 17.4533 1 0.0014 Contact Force 195.724 1 0.0209 Base Power 0.63268 1 0.3643 Base Time 48.316 1 0.1840 Base Force 0.00550239 1 0.9304 Base Power 68.254 1 0.1233 Gas Flow 17.72 1 0.0013 Base Force 86.8062 1 0.0888 Scrub 4.15493 1 0.0024 Gas Flow 22.2414 2 Error 0.672107 1 0.0418 Scrub 0.476631 5 Replicate Error 17.4533 2 Error Table 6: ANOVA table for response: Yield. 0.635 5 Replicate Error

Response: Pull test minimum value Table 5: ANOVA table for response: Pull test minimum value Degrees Mean of P Control Factors As regards to Pull test average and Pull test minimum Squares Freedom value, experiment can be described as successful /1/. 10.2541 1 0.0075 Standby Power Causes of quality characteristics variations are detected, 31.1104 1 0.0003 Contact Time by experiment results analysis. 1.31813 1 0.2248 Contact Power For response Yield, there is a lack-of-fit /4/; measured 11.8714 1 0.0052 Contact Force values don’t fit to the theoretical design model. That 0.08366 1 0.7472 Base Time means that response are affected by some other factors 3.69629 1 0.0610 Base Power which were not included in experiment, or assumed 32.3320 1 0.0003 Base Force linear model is not correct – correlations between factors 7.2044 1 0.0170 Gas Flow should be included /1/. 1.54401 1 0.1928 Scrub 0.74373 2 Error 0.4 5 Replicate Error Table 4: ANOVA table for response: Pull test average value. 3 IDENTIFYING A PROCESS PARAMETER In the third coloumn (tables 4, 5 and 6) P- values (F- WINDOW test estimated values) for control factors are shown. By dividing the variance (sum of squares divided by the Contour plots are principal means for displaying number of degrees of freedom) due to each factor by the relationships between response (quality characteristic) variance of the error, a measure of the relative and continuous control factors (parameters). With the aid significance of each factor is obtained (F-distribution) /3/. of a graphical contour plot , it is possible to define the The P-values are upper tail values from F-distributions. optimized parameter window based on the target As such, values less than 0.05 - for 99.9% confidence, response specification. This technique can show how is 0.01 - for 99.9% confidence, and 0.005 - for 99.9% particular response affected by given set of control confidence, are considered as statistically significant. /4/. factors over some specified region of interest, and what Considering P – values from table 4, significant settings of factors will give a product simultaneously factors for response Pull test average value, are satisfying desired specifications. Optimization can be respectivly: obtained by localization of a quality characteristics - Standby Power (with 99,9% confidence), optimum area on the contour plot. There are three - Contact Time (with 99,9% confidence), responses in this experiment. Combined Response - Base Force (with 99% confidence), technique will be used to define process parameter - Contact Force (with 99% confidence), window. - Gas Flow (with 99% confidence), - Scrub (with 95% confidence). 3.1 Combined Responses . Significant factors for response Pull test mimum Multiresponse optimization use superposition of value, with regards to their P – values, are respectivly: contour plots for all responses, to find the specific area - Contact Time (with 99,9% confidence), on the superpositioned plot, which will satisfy - Base Force (with 99,9% confidence), specifications for all responses. A combined response is a - Contact Force (with 99% confidence), response variable created as weight sum of several - Standby Power (with 99% confidence), individual responses. - Gas Flow (with 95% confidence). Let xi, i=1, …., p be a set of response values, then a combined response is defined as /4/:

p

97 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 g ∑ ki wi │(xi – gi) / ri│, (1) OPTIMUM 1 20

where ki is 1 when response i is to be maximized or .6 3 minimized, and -2 otherwise; wi is a user chosen weight .4 3 for response i; and ri is the range of xi . The values of gi, .2 15 3 depends on whether the response is maximized, .0 3 minimized or targeted. If the response is targeted, the gi .6 is the targeted goal; if it is maximized than gi = xi – ½ ri ; 2 Contact_Time if it is minimized than g = x + ½ r ; where x and x are i i i i i 10 .2 2 the minimum and maximum of xi. .0 2 Figure 6 shows spreadsheet for Combined Response: .8 OPTIMUM, where all three responses are set to 1 ECHIP maximum – maximum pull test values and maximum 5 process yield are requested. 10 20 30 40 50 Due to the fact that, by quality specification all pull Standy_Power test values measured in one device have to be higher than lower specification limit, weight factor for response Pull Figure 7: Contour Plots for Combined Response OPTIMUM Test minimum value is set to 2, and for other two against Standby Power and Contact Time. g responses weights are set to 1. OPTIMUM

150

.2 3

.0 3

100 Contact_Force

.4 2

.2 2 ECHIP 50

20 30 40 50 60

Contact_Power Figure 6: Spreadsheet for Combined Response OPTIMUM. Figure 8: Contour Plots for Combined Response OPTIMUM At the following figures, contour plots for combine against Contact Power and Contact Force. response against pair of factors found as significant in ANOVA tables will be shown, excluding factor Scrub Base Time and Base Power effects on combined SPC with categorical values. response OPTIMUM are shown at figure 9. Since these Numbers shown at the plot lines show relative two factors are insignificant for pull test values, and probability to satisfy all conditions defined by combined significant for response Yield but there is lack-of-fit for response. That means that higher number assigned to the response Yield, this graph will not be analyzed. line shows higher probability to reach combined Figure 10 shows Contour Plots for Combined response. Response OPTIMUM against Base Power and Base In figure 7, Contour Plots for Combined Response Force. Factor Base Power unsubstantially affects OPTIMUM against Standby Power and Contact Time, combined response. Maximizing Base Force value will one can see that minimum value for Standby Power and positively affect combined response. maximum value for Contact Time is the best combination In figure 11, Contour Plots for Combined Response to reach conditions described in combined response. OPTIMUM against Base Force and Gas Flow, it is Figure 8 shows Contour Plots for Combined visible that maximum value for both factors is favorable. Response OPTIMUM against Contact Power and Contact Force. Choosing minimum value for Contact Power and maximum value for Contact Force, desirable combined response can be obtained.

98 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Outside Design OPTIMUM - Contact Power = / - Contact Force = 150 gram, 120 - Base Time = 40 ms, - Base Power = 80 mu,

110 - Base Force = 400 gram, - Gas Flow = 1 l/min, - Scrub SPC = “Yes”. 100 Since scrubbing will be applied (Scrub SPC = “Yes”), Base_Power values for Standby Power and Contact Power are set to 0.

90 It would be favourable to further maximize values for factors Contact Time and Base Time, but then process would be long lasting, and that would affect bare copper ECHIP 80 oxidation. It has been decided that maximum values used 20 25 30 35 40 in experiment are adopted. For factor Contact Force, maximum value used in Base_Time experiment is adopted. Figure 9: Contour Plots for Combined Response OPTIMUM Due to the fact that Base Force is one of the most against Base Time and Base Power. significant factors for pull test values, chosen value for Base Force is out of the range used in experiment.

g For Base Power, minimum value used in trials is OPTIMUM adopted – 80 mu.

350 Chosen Gas Flow value is set to maximum level that 3.2 machine instrument can measure - 1 l/min. Using special module, ECHIP can calculate possible 3.0 responses and their limits, for specific set of control factors. For afore chosen factors values, calculated 2.8 responses are: 300

Base_Force • Pull test average value = 44.78 grams (limits: 42.62 – 46.94 grams) • Pull test minimum value = 30.32 grams 2.4 (limits: 28.05 – 32.59 grams). ECHIP

80 90 100 110 120

Base_Power 4 VERIFICATION – VALIDATION RUN

Figure 10: Contour Plots for Combined Response OPTIMUM For chosen parameters values, another run is against Base Power and Base Force. performed to validate the long-term stability of process.

g Using adopted parameters window, validation run is OPTIMUM performed, under the same conditions as experiment. Sample size was 120 devices. After performing wire 1.0

3 bonding, pull test values have been measured on 6 . 4 randomly chosen devices (table 7).

3 Pull test average value Pull test minimum . Device No. 2 (gram) value (gram) 3 . 0.8 0

Gas_Flow 1 2 45.43 31 . 4

2 44.12 30

2 3 . 45.84 30 2

ECHIP 0.6 4 45.63 31 300 350 5 44.53 30 Base_Force 6 46.57 30 Figure 9: Contour Plots for Combined Response OPTIMUM against Base Time and Base Power Software 44, 78 30,32 generated (42,64 – 46,94) (28,05 – 32,59) From ANOVA analysis and contour plots for values combined response, adopted process parameters values are: Table 7: Sample from verification run - Pull test values - Standby Power = / measured on 6 devices. - Contact Time = 20 ms,

99 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Results show that Pull test average value range (44.12 In these 2 figures, it is visible that all welds are – 46.57 gram) is within software calculated limits (42.64 visually acceptable, with no splits or rifts. – 46.94 gram), for the same parameter values. For Pull test minimum value, results of measurement (30 – 31 gram) are with limits predicted by software (28.05 – 32.59 gram). 5 CONCLUSION These indicate that variations for Pull test average and Pull test minimum values are under control /1/. Using Taguchi Design of experiment model, robust Furthermore, one can conclude that robust process process “copper wire bonding to bare copper leads” has “copper wire bonding to bare copper leads” is developed, been developed /1/. Products – microelectronic devices as regards to pull test values /1/. satisfy specifications for quality characteristics. During verification run, in the 120 devices sample Advantages of combined approach using Taguchi where every device contains 58 wires, 5 copper wires technique and contour plots for combined response, for were unsuccessfully welded to the bare copper leads. redesign of thermosonic copper wire bonding process, During verification run, in the 120 devices sample are: where every device contains 58 wires, 5 copper wires - Required number of trials, to design process, is were unsuccessfully welded to the bare copper leads. minimal. In this experiment with 9 control factors, only Process Yield calculation is: 12 trials was performed (with 5 repetition which are not - number of wires in the sample: mandatory) instead of 2048 trials using classical full 120 devices * 58 wires = 6960 wires, factorial approach; - process Yield: - There is possibility to adopt economical ( (6960 – 5 ) / 6960 ) * 100 % = 99,928 %. sequential approach for experiment performing; Process yield obtained during verification run is - For parameters values chosen from contour satisfactory, considering the fact that this is experimental plots, it is possible to predict response - quality / researching phase of “copper wire bonding to bare characteristics value; copper leads” process. - In general, possibility to optimize process and Figure 12 shows few copper wires welds at bare improve process robustness. copper leads (magnification 10 x) /1/. The use of Taguchi experimental design techniques was successful; it resulted with development of a robust process - copper wire bonding to bare copper leads, with no silver or any other material as mediator /1/. So far, information about successfully performed thermosonic bonding process with copper wire to bare copper leads, were not found. copper wire Beside process cost reduction caused by elimination of silver from microelectronic device, device quality is weld improved, due to favorable electrical and thermical characteristics of copper /1/.

bare copper lead surface REFERENCES

Figure 12: Copper wire welds at bare copper leads /1/ Sibalija, T., 2004, Application of Taguchi Model for (magnification 10 x). Quality Product Improvement, Master thesis (in preparation), Belgrade, Serbia&Montenegro. Cross-section of copper wire weld at bare copper lead /2/ Various Authors, 1999, AB339 Gold Wire Ball is shown at figure 13 (magnification 40 x) /1/. Bonder – Technical Operator Training Material, ASM Technology Singapore Pte.Ltd. /3/ Wilson, G.,1994, On Route to Perfection – A Guide to Achieving Six Sigma, IFS International Limited. /4/ Wheeler, B., ECHIP Software for the Design of copper Experiments – Reference Manual, ECHIP Incorporated, US. wire

weld

copper lead

Figure 13: Cross-section of copper lead with copper wire weld (magnification 40 x).

100 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

USING PARTIAL LEAST SQUARES REGRESSION TO ANALYSE QUALITY IN HIGHER EDUCATION

Keynote paper L. Catellani, Bianca M. Colosimo∗, Q. Semeraro Dipartimento di Meccanica – Politecnico di Milano, Milano (Italy)

Summary: The application of the EFQM Excellence Model to Higher Education gives the opportunity to measure and improve the effectiveness and the efficiency of Educational Organizations. The basic principle of EFQM Model is the customer satisfaction. With reference to Higher Education Institutions, many actors play the role of customers: students, faculty and administrative employers, companies involved in research projects, etc. Among these actors, students have a relevant role in judging services provided by a university: collecting and analysing opinions about their satisfaction should be interesting and helpful. Questionnaire is the most used tool to measure students’ satisfaction. It contains some questions concerning different aspects of the service (i.e., services’ attributes) with regards to organization, infrastructures, teaching methods and teacher skills, course content and global satisfaction. For each attribute, students can express their opinions through a Likert scale (agree/not agree gradient) that is a scale with numeric labels for assessing satisfaction. To understand how students’ requirements can be met and satisfaction improved, data collected through questionnaires have to be analysed to derive a model that specify the relative importance of each attribute in creating the overall satisfaction. In this case, traditional regression analysis can not be adopted because of multicollinearity. In fact, in most real cases answers related to different questions are highly correlated. To overcome this problem, this paper presents an approach to analyse the collected data on students’ perceptions. The approach is based on using Partial Least Squares (PLS) Regression and is applied to data on students’ satisfaction collected for courses offered in three semesters in a school of engineering. In particular, collected data refer to a total amount of 599 courses evaluated by 27687 students in three semesters. The results explain that factors related to teacher’s ability are the most important elements that affect satisfaction. Key words: Excellence, Improvement, Customer Satisfaction, Education

1. QUALITY IN HIGHER EDUCATION EFQM model situates the client as the focus and as the INSTITUTIONS main reason for all the activities of the organisation //2//. Orienting organisations to customers means that policy Total Quality Management has been used to support a and strategy of the organization should be oriented to quality improvement process. A recent model that applies meet actual expectations and future needs of the client. In the TQM philosophy is the EFQM (European Foundation this contest is important to do a clarification about the for Quality Management) Excellence Model. concept of customer. Proposed in 1991 with the aim of making ‘European Customers would be external or internal. External Business more competitive’, nowadays the EFQM Model customers are the final users of products and services is the reference model for those organisations whose aim offered by the organisation. Internal customers are is achieving “excellence”. Hence, applications of the suppliers, partners and all the people who have economic EFQM Model are aimed at increasing the interests in the organisation, i.e., people that are competitiveness and the effectiveness of the organisation integrated within the organisation. and at enhancing quality of provided services. The EFQM In Higher Education Institutions (HEI) internal does not focus on standardising quality systems, instead it customers are faculties and administrative people while defines a management tool to promote excellence in external customers are students, companies cooperating quality management, for instance through the European with university for research purposes and, from a wider Quality Award Scheme. The EFQM Model can be in point of view, the society in which the institution particular used as a tool for self-assessment, as a way to operates. benchmark with other organisations, as a guide to identify The EFQM Excellence Model is a non-prescriptive areas for improvement //1//. In fact, an organisation can framework based on 9 criteria: five of them refer to regularly evaluate itself by using a set of performance 'Enablers' and the remaining four ones are 'Results'. The criteria with the aim of achieving quality improvement. 'Enabler' criteria cover what an organisation does. The In this paper, we will focus on a particular type of 'Results' criteria cover what an organisation achieves. organisation: a Higher Education Institution. When 'Results' are caused by 'Enablers' and 'Enablers' are EFQM’s principles are applied to improve quality in this improved using feedback from 'Results'. In this paper we type of organization, the first problem is defining will focus on results obtained by an Higher Education customers. In fact, one of the fundamental excellence Institution with respect to a specific process and a concepts in EFQM regards customer focus. In this sense, specific customer: the teaching process and the students.

∗ Corresponding author 101 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Compared with the previous study presented by Chua can be made for students’ benefit. Finally, the end-of- //3// to analyse quality for higher education, our analysis term questionnaire is often preferred to have a complete has a different goal. Chua proposed an ‘ad hoc’ evaluation of student satisfaction, because questions are questionnaire submitted to small samples of students asked when all the learning activities have been (35), parents (27), faculty (10) and firms (12) to completed and students can thus give meaningful answer understand how quality in high education is perceived to questions about the course //8//. from different “customers”. He distinguished three main Questionnaires adopted have often a general structure attributes of quality in this framework: ‘Input’, which in which the questionnaire is organised into sections refers to the entry requirements, ‘Process’, which refers related to different characteristics: the organisation of to the teaching and learning process, and ‘Output’ which teaching activities; the evaluation of teacher’s skill (e.g., refers to the employability and academic standings. The clearness, effectiveness, ability to communicate interest objective of the analysis was to understand how quality is on the topic); the evaluation of infrastructures perceived by different customers. (concerning class and laboratories) and the overall In our study we will focus on a specific process assessment of the global satisfaction related with the (teaching activity) and a specific customer (the student) specific course. For each question, the student is usually to understand which are the main aspects that influence asked to express a judgement concerning his/her his/her satisfaction. To accomplish this goal, we satisfaction with reference to an ascending four points considered a large sample size of questionnaires. In fact, Likert scale. Conventionally, Likert scale is a closed we analysed data related to three semesters, compiled by response scale with ordinal numeric labels for assessing a total amount of 27687 students and related to a whole satisfaction; it is employed to measure attitudes or set of 599 courses. opinions and it generates ordinal variables that are often converted in numbers (i.e. the typical question is “Are you satisfied with this specific aspect of the course?” where possible answers are: 1- Definitely not; 2- No, 2. STUDENTS’ SATISFACTION rather than yes; 3- Yes, rather than no; 4- Definitely yes). This translation of ordinal variables into numerical Quality of service provided is often related to outcomes, which is called “scaling”, has been discussed customer satisfaction. According to Juran //4//, service in the literature. As outlined in //7// scaling assumes quality can be defined as the extent to which the service location of attitudes on a continuum and equidistance successfully fulfils the needs of a customer, i.e., his/her between alternatives. Different approaches have been satisfaction. Although some authors outlined that service proposed in the literature to perform service quality quality and customer satisfaction should be considered as analysis on ordinal variables when scaling is not different concepts, most of the studies on service quality performed //9/,/10//. However in this paper we assume reported that these two concepts are highly correlated scaling is already performed because the possible answers //5// and, besides, indistinguishable for some customers for the student are already expressed in a scale ranging //5//. from 1 to 4. In this analysis we will focus on assessing a As previously mentioned, a typical questionnaire is methodology to investigate aspects related to student’s composed by a set of questions related to specific satisfaction. Student satisfaction represents one of the characteristics of the service provided and a last question performance indicator that should be constantly observed related to the overall satisfaction. This last question can to find improvement keys, according to the EFQM be seen as a synthetic judgement of the perceived quality. philosophy. Therefore the analysis of data collected should be aimed There are many tools for assessing students’ at analysing to what extent each specific characteristic satisfaction: questionnaires, open interviews, focus (represented by all the questions but the last one) groups, and external monitoring //7//. In Higher influences the global satisfaction (related to the last Education Institutions it is usual practice to collect question). In other words, the students satisfaction’s key through questionnaires students’ feedback about the factors should be identified and their importance perceived quality. estimated. This paper presents an approach that uses A main advantage of using questionnaires is that they Partial Least Square Regression to perform this analysis. allow anonymous (and therefore probably truthful) answers. A limitation is instead related to the reduced number of possible answers (usually related to a scale that ranges from “extremely unsatisfied” to “extremely 3. ANALYSIS OF DATA COLLECTED WITH satisfied”). In fact, this rigid scale does not allow to QUESTIONNAIRES collect specific explanation of reasons behind judgements. Questionnaires can be given at the In order to describe the relationships among students’ beginning, in the middle and at the end of a course. Some overall satisfaction and the other specific attributes teachers use questionnaires at the beginning of a course to related to teaching activities (different questions in the get information about the students (prior course or questionnaire), standard Multiple Linear Regression experience with the subject, preferred forms of teaching (MLR) could be used. In fact, MLR allows to find and learning). Many universities use mid-term dependencies between a response variable and a set of questionnaires to allow for an early warning on existing regressors or predictor variables X j (1,...,)jJ= . In this problems in order to decide proper feedback actions that 102 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 framework, the overall satisfaction plays the role of the correlated. As in Multiple Linear Regression, the main response variable while the other aspects of teaching can purpose of PLSR model is to estimated a linear model: be seen as predictor variables. Y=XB+E (2) Unfortunately, a main assumption behind MLR is where Y is an n cases (observations) by m variables independence among predictor variables. In fact, when response matrix; X is an n cases by p variables predictor correlation among these variables is strong, matrix, B is a p by m regression coefficient matrix, and multicollinearity arises. Multicollinearity can have E is a noise term for the model which has the same serious effects on the estimates of the regression dimensions as Y . Usually, variables in X and Y are coefficients and on the general predictive ability of the centred by subtracting their means and scaled by dividing model derived. The presence of multicollinearity can be by their standard deviations. detected in several different ways, in particular: //11// To find the model in (2), PLSR transforms the set of I. Bivariate correlation between predictor variables. If correlated explanatory variables X into a new set of there is high correlation between coefficients related uncorrelated variables T , where: to predictor variables, then multicollinearity is T=XW, (3) suspected. where T is a n by c matrix called factor score matrix, II. Variance Inflation Factors (VIF) associated with the W is a p by c matrix called weight matrix. These predictors variables. The VIF for the jth regression weights are computed to maximise the covariance variable is defined as: between responses and the corresponding factor scores in 1 T . Ordinary least squares procedures for the regression VIF = (1) j 2 of Y on T are then performed to produce Q , the 1− R j loadings (or weights) for Y such that 2 where R j in (1) is the coefficient of multiple Y=TQ+E. (4) determination obtained from regressing X j on other where E is an error (noise) term. By substituting (3) in (4) we have (2) where B=WQ, and the prediction regressor variables model is complete //15//. X = X ,XXX ,..., , ,..., X. The higher the {}j {}12jj−+ 1 1 J The number c of factors in T that are included in the VIF, the more severe is the multicollinearity. As a rule of regression model can be selected by using cross- thumb, to avoid multicollinearity, VIF should not exceed validation. Cross validation is based on measuring the 4 or 5 //11//. prediction ability of competing models. In this case, Several remedial measures have been proposed for competing models are characterised by a number of solving multicollinearity. A possibility is to delete certain factors ranging from 1 to p. In particular, the leave-one- variables from the model, but this approach has the out version of the cross validation algorithm is widely disadvantage of discarding the information contained in used and is described in the following. the deleted variables //11//. To deal with Each competing model is computed as many times as multicollinearity, Partial Least Squares Regression can be there are observations, omitting a different observation used. each time. For each omitted observation, the fitted or predicted response using this model is calculated. The 3.1. Partial Least Squares Regression – PLSR difference between this predicted value and the actual observed response value is called prediction error. Once Partial Least Squares Regression (PLSR) originates in this routine is repeated for all the observations, the 1975 in a study by Herman Hold related to extracting prediction sum of squares (PRESS) is computed as the relevant information from data produced in chemical sum of squares of the prediction errors. In general, a experiments. For many years PLSR was related to smaller PRESS values indicates a better prediction ability Chemometrics studies. Chemometrics is the chemical of the model. Therefore the competing models discipline that uses mathematics and statistic to design characterised by a smaller value of the PRESS statistic is optimal measurement procedures and experiments, in selected. order to provide the most relevant information in the To evaluate how well this selected model predicts analysis of chemical data. In particular, PLSR found responses for new observations, R 2 can be computed many applications in Multivariate Calibration, where the pred goal is to build a multivariate model to predict a chemical as: or physical properties from a set of predictors variables 2 PRESS Rpred =−1 . (5) //12//. SStot During the last decade, PLSR’s applications begin to 2 where SS is the total sum of squares. R is spread in contexts different from Chemometrics. As an tot pred 2 example, PLSR is recently suggested for multivariate between 0 and 1 and larger values of Rpred suggest quality control in manufacturing //13/,/14//. models of greater predictive ability //12//. PLSR is a technique that generalises and combines features from principal component analysis and multiple regression. It is a multivariate data analysis technique that can be used to relate one or more response variables (Y ) to several explanatory variables ( X ) that can be

103 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

4. THE CASE STUDY a survey must be carried out at the end of each course. In this work a real case has been studied, where collected In Italy, the evaluation of university’s teaching data arise from an Italian school of engineering. Table 1 activities became a process established by a national Law presents the questions included in the questionnaire used in 1999. Since this date, to obtain the evaluation of the in this research students with respect to the different aspects of a course, .

X1 Is the overall loading of study acceptable in this semester? A – Organisation of the whole semester Is the overall organisation (working hours and middle/final examination) X 2 acceptable in this semester?

X3 Is the grading system (exams’ procedures) clearly defined? B – Organisation of this specific course X4 Are the schedules respected? X5 Is the teaching staff available for explanations? Is my preliminary knowledge adequate for understanding the topics tackled X 6 in this teaching?

X7 Does the teacher stimulates/motivates the interest? C – Teaching activities and study related X8 Does the teacher explains the arguments clearly? to the course X9 Is the required workload proportional to credits assigned? X10 Is the didactic material (suggested or supplied) adequate? Are the supplementary activities (workshops, laboratories) helpful for X 11 learning? Are the classrooms adequate (i.e., it is possible to see, to hear and to take a X 12 seat)? D - Infrastructures Are the locals and the equipment used for supplementary activities X 13 (workshops, laboratories) appropriate? Am I interested to this course’s topics? (independently on how teaching has X E – Interest and satisfaction 14 been performed) Y Am I overall satisfied concerning this course?

Table 1 – The set of questions There are five macro areas (A, B, C, D, E) of provided. Considering the model described in (2), investigation, where questions are specifically related to Y=XB+E , in this case Y is an 193 cases different aspects of the service with regards to semester’s (observations) by 1 variable response vector; X is an organisation, course’s organisation, teaching activities, 193 cases by 14 variables predictor matrix, E is a 193 infrastructures and global satisfaction. For each question, by 1 noise vector and B is a 14 by 1 regression four possible answers are foreseen: coefficient vector. The aim of our analysis is to compute 1- Definitely not; B . In fact, the j-th element bj in this regression 2- No, rather than yes; coefficient vector can be interpreted as the importance 3- Yes, rather than no; weight of the j-th attribute ( j = 1,...,14 ) on the overall 4- Definitely yes. The questionnaire form contains 15 items where the satisfaction. last question is the question that summarises the overall satisfaction. 4.1. Results Data collected concern 193 courses attended by a As described in Section 3, a first investigation of data whole amount of 8245 students in the second semester of collected concerns multicollinearity analysis. In fact, if the academic year 2003/2004. predictor variables are independent (i.e. no For each course, students who attended that course multicollinearity is present) simple Multiple Linear express a judgement (ranging from 1 to 4) for each Regression (MLR) could be used instead of PLSR. question and the sample mean of these judgements is With reference to data collected in the case study, collected. Therefore, the generic element xij of the Table 2 reports Pearson correlation coefficients between matrix X in (2) is the mean value of judgements each pair of predictor variables (and - in parenthesis - the expressed by students in the i-th class for the j-th associated p-values for the hypothesis test of the specific question, where i = 1,...,193 and j = 1,...,14 . correlation coefficient being zero). The last row in Table Finally, the response vector Y in (2) is a 193 by 1 2 reports the VIF values associated with each variable in vector, where y is the mean value of judgements X . As it can be observed, most of the correlation i coefficients are significantly different from zero1, i.e., concerning the overall satisfaction for the i-th course ( i = 1,...,193 ). 1 The goal of our analysis is to derive the model which To test the assumption that all the (14)(14-1)/2=91 correlation coefficients are all equal to zero, the Bonferroni inequalities can relates the overall satisfaction (Y ) to the satisfaction be used. Assuming a first type error equal to 10%, p-values related to each specific attribute of the teaching service shown in parenthesis can be compared with α /91= 0.0011. 104 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 there is high correlation between variables in X . PLSR has to be used to overcome the multicollinearity Furthermore, many VIF’s are greater than 4 and hence problem.

Table 2 – Correlation coefficients matrix and VIF values

X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X 0.831 2 (0.000) X 0.388 0.481 3 (0.000) (0.000) X 0.163 0.335 0.55 4 (0.023) (0.000) (0.000) X 0.357 0.491 0.715 0.674 5 (0.000) (0.000) (0.000) (0.000) X 0.462 0.461 0.375 0.213 0.451 6 (0.000) (0.000) (0.000) (0.003) (0.000) X 0.352 0.478 0.645 0.501 0.795 0.507 7 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) X 0.368 0.464 0.614 0.475 0.704 0.541 0.834 8 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) X 0.749 0.617 0.479 0.316 0.477 0.598 0.467 0.495 9 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) X 0.407 0.477 0.691 0.507 0.627 0.382 0.561 0.621 0.534 10 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) X 0.419 0.522 0.567 0.403 0.566 0.344 0.527 0.488 0.498 0.569 11 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) X 0.23 0.342 0.252 0.172 0.267 0.236 0.23 0.249 0.188 0.323 0.271 12 (0.001) (0.000) (0.000) (0.017) (0.000) (0.001) (0.001) (0.000) (0.009) (0.000) (0.000) X 0.242 0.28 0.248 0.191 0.289 0.155 0.257 0.276 0.25 0.384 0.319 0.793 13 (0.001) (0.000) (0.001) (0.008) (0.000) (0.032) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) X 0.248 0.329 0.45 0.251 0.502 0.501 0.651 0.468 0.325 0.402 0.416 0.164 0.165 14 (0.001) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.023) (0.023) VIF 5.7 4.7 2.7 2.1 4.4 2.2 4.3 3.7 2.8 1.9 3.1 3.1 2.1

Table 3 shows the regression coefficients bj Examining PLSR coefficients, it can be observed that ( j = 1,...,14 ) obtained using PLSR (where the model service’s attributes related with questions X8 (teacher’s clearness) and X (teacher’s ability in motivating estimated has R 2 equal to 0.876). Table 3 reports in the 7 pred interest) appear as the most important key factors in last column the correlation coefficient between each creating the overall satisfaction. Clearness in defining the variable in X and the response variable (Y ) that grading system (exams’ procedures), i.e. X , is also represents the overall satisfaction. 3 perceived as important. Furthermore, questions regarding Ar semester organisations ( X1 and X 2 ) and infrastructures Variable PLSR coeff Corr coeff ea ( X12 and X13 ) are almost insignificant in determining X 0.0126 0.512 A 1 the overall satisfaction. X2 0.0505 0.606 To outline advantages related with PLSR, consider X3 0.1575 0.772 the effect related with question X5 (Is the teaching staff B X4 -0.0086 0.544 X5 0.0808 0.803 available for explanations?). This attribute presents a X6 0.0065 0.580 high correlation coefficient (0.803) with the overall X7 0.1905 0.858 satisfaction but a low weight (0.0808) after PLSR is X 0.2796 0.858 C 8 performed. This mean that the correlation coefficient X9 0.1376 0.657 (0.803), that seems to assess a direct relationship between X 0.1386 0.729 10 X5 and the overall satisfaction, is just representing a X 0.1457 0.671 11 hidden correlation between this variable and some other X -0.0018 0.280 D 12 variables (i.e., variable X ) that highly affect the overall X13 -0.0453 0.299 3 E X14 0.1101 0.618 satisfaction.

Table 3 – PLSR regression coefficients and correlation coefficients (between each variable and the overall satisfaction’s one) .

105 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

5. EXTENDING THE ANALYSIS ƒ the teacher’s ability in motivating students’ interest ( X7 ) and To assess whether conclusions derived were just ƒ clearness in defining the grading system related to a specific semester or should be considered (exams’ procedures) ( X3 ); stable with time, the PLSR analysis was repeated going while attributes regarding infrastructures are always back on data related to the previous two semesters. In irrelevant ( X and X ) are almost insignificant. particular, we considered data related to 221 and 185 12 13 courses offered in the first semester of 2003/2004 and in the second semester of 2002/2003, respectively. In these two semesters the same questionnaire’s form was compiled by a total amount of 9918 (first semester 6. CONCLUSIONS 2003/2004) and 9524 (second semester 2002/2003) students. This paper presented an approach to rank factors Results on PLSR regression coefficients are affecting student’s satisfaction. It has been based on represented in Figure 1. adopting Partial Least Squares Regression to analyse data As it can be observed, in all the three semesters, the concerning satisfaction for courses offered in three most important aspects are related to: semesters in an Italian school of engineering. ƒ the teacher’s clearness ( X8 );

II semester (03/04) 0.30 I semester (03/04) 0.25 II semester (02/03)

0.20

0.15

0.10

0.05

0.00

-0.05 X8 X7 X3 X11 X10 X9 X14 X5 X2 X1 X6 X12 X4 X13

Figure 1 - Regression coefficients in the last three semesters

The approach presented allows to assess the central The approach can be also used to re-design role of teachers in determining students’ overall questionnaire. In fact, attributes that showed a small satisfaction. In fact, students are basically satisfied when impact on the overall satisfaction could be eliminated the teacher explains clearly and motivates interest. while more than one questions should be dedicated to Furthermore, students give relevance to a clear definition better investigate key-factors outlined by the analysis. of the course’s grading system. A further remark concerns the applicability of the Obviously, these results should suggest proper actions approach proposed in different contexts. In fact, the use aimed at improving students’ satisfaction for these key- of PLSR for customer satisfaction analysis is applicable factors, in a continuous improvement framework. The any time questionnaires are used to collect data and ways in which these actions should be identified and possible answers are based on a numeric Likert-type implemented is strongly dependent on the specific scale. context in which the university operates and this issue goes beyond the scope of this paper. The main goal of the present research was hence to present a quantitative and objective approach for REFERENCES analysing data related to quality perceived by students. The approach proposed can be easily applied as a support /1/ European Foundation for Quality Management: Radar tool by any faculty interested in identifying key-factors and the EFQM Excellence Model. EFQM,1999 affecting the overall satisfaction of students. http://www.efqm.org/

106 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

/2/ Martin-Castilla J.I., Possible Ethical Implications in the Deployment of the EFQM Excellence Model. Journal of Business Ethics. 39, 125–134, August 2002. /3/ Chua C., Perception of Quality in Higher Education, AUQA Occasional Publication Proceedings of the Australian Universities Quality Forum, 2004. /4/ Juran's Quality Handbook (5th Edition) Edited by: Juran, J.M., McGraw-Hill, 1999. /5/ Sharma S., Niedrich R.W., Dobbins G. A Framework for Monitoring Customer Satisfaction: An Empirical Illustration. Industrial Marketing Management. 28 , 231–243, 1999. /6/ Dabholkar, P.A., Shepherd, C. D., Thorpe, D.I. A Comprehensive Framework for Service Quality: An Investigation of Critical Conceptual and Measurement Issues Through a Longitudinal Study. Journal of Retailing. 76 (2), 139–173, 2000. /7/ Franceschini F. Dai prodotti ai servizi. UTET. 2001. /8/ Dee Fink L., Evaluating Your Own Teaching, Improving College Teaching, Peter Seldin ed., 1995. /9/ Colosimo B.M., Semeraro Q., A Bayesian control chart for service quality control, Quality & Productivity Research Conference, Tempe, Arizona June 5–7, 2002. /10/ Franceschini F., Galetto M., Varetto M. Ordered Samples Control Charts for Ordinal Variables. Quality and Reliability Engineering International, 21 (2), 177–195, 2005. /11/ Montgomery D.C., Peck E.A., Introduction to linear regression analysis. John Wiley & Sons. 1992. /12/ Martens H., Næs T., Multivariate Calibration. John Wiley & Sons. 1989. /13/ Kourti T., MacGregor J.F., Multivariate SPC Methods for Process and Product Monitoring. Journal of Quality Technology. Vol. 28, n. 4, October 1996. /14/ Wurl R.C., Albin S.L., Shiffer I. J., Multivariate Monitoring of Batch process startup. Quality And Reliability Engineering International, 17, 269–278, 2001. /15/ Wold S., Sjostrom M., Eriksson L. PLS - regression: a basic tool of chemometrics. Chemometrics and Intelligent Laboratory Systems, 58, 109–130, 2001.

107

Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

RESEARCH AND DEVELOPMENT OF DIGITAL QUALITY MODEL IN SCM Keynote paper 1) Nenad Stefanovic, Zastava Automobiles, 2) Vidosav Majstorovic, 3) Dusan Stefanovic, 1)Information Systems Division, Kragujevac, SCG 2)Mechanical Engineering Faculty, Belgrade, SCG; 3)Faculty of Science, Kragujevac, SCG

Summary: Quality initiatives and reengineering were forcing companies to evaluate entire processes, rather than individual components. Supply Chain Management (SCM), the integrated control over goods, information, and money, followed. SCM deals with total business process excellence and represents new way of managing the business and relationships with other members of the supply chain. This collaborative business network necessitate the new digital quality management model that enables delivering high quality products and services in the right place, in a timely fashion, and cost effectively. This paper discusses the needs for digital quality model in SCM, analyses existing quality models, and purposes the framework facilitated with modern information technologies for achieving this goal. Keywords: SCM, Digital Quality, Modelling.

1. THE NEW BUSINESS PARADIGM Internet has contributed to both the increasing needs and opportunities for improved supply chain management. Today’s businesses face many challenges on how to With the Internet, companies in a supply chain can be optimally run their organizations. Dynamic economic connected in real time with information and knowledge conditions, stretched resources, and shrinking budgets shared continuously, new products and services can be lead to a strong focus on ensuring that there is a return on designed to fit special market segments, and new supply every dollar spent. Customers now demand custom- chain structures can be developed to serve customers in a tailored products, personalized services, and desired more direct manner [1]. quality level. The use of modern information Many companies are beginning their search for a technologies and rapid prototype development are solution to implementing an electronically oriented approaches that fulfil those demands according to digital supply chain management system that provides factory concepts. connection to customers and to suppliers. This integrated Digital factory integrates generating, storing, and supply chain may be based upon new software solutions delivering of all relevant data about products (services) or based upon enhanced communication capabilities. The and processes throughout life cycle – from early planning ultimate objective is to create a “seamless system to mass production and returns management. Digital interface” that provides the capability to review and factory concept supports the model of cooperative analyze varying elements of information. The objectives engineering. It unifies digital product or service, digital for analysis of this information are to create a more planning, digital operations, and real factory. efficient supply chain characterized by: In today’s global competitive market, first class • Increased customer service levels; products no longer guarantee success in the aggressive • Decreased transaction costs; battle for market share. To be successful, organizations • More efficient inventory investments; need to increase their productivity and reduce costs. To • Reduced expenses for manufacturing; survive in the present global competitive environment, • Increased responsiveness to customer demands; organizations need to show a heightened awareness to • The ability to fulfil customer requirements more customers needs. Hence, there is an increased focus on a profitably; customer-centric business models and integration of • The ability to deliver high quality products in supply chains to enable collaboration between the supply the shortest time; chain partners. • The ability to deliver products at the lowest cost; Supply chain management has emerged as one of the • The ability to penetrate smaller, fragmented major areas for companies to gain a competitive edge. markets cost effectively; Supply chain management is the process of developing • Greater linkages with key suppliers; decisions and taking actions to direct the activities of • Demand driven supply chain; people within the supply chain toward common • Capacity planning across the supply chain; objectives. It is seen as a tool for managing resources and • Sharing of information with key suppliers thus optimizing product and material flows across the network reducing supplier costs. of supply chain participants. In order to achieve these objectives we need a supply Managing supply chains effectively is a complex and chain-wide digital quality management system that will challenging task, due to the current business trends of provide consistent framework for establishing, managing, expanding product variety, short product life cycle, measuring, and improving supply chain processes. This increasing outsourcing, globalization of businesses, and system should be integrated into the overall digital continuous advances in information technology. The factory framework. With its feedback mechanism, quality

109 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 management ensures a coordination and synchronization delays and problems caused by changes in the extended among digital factory and real factory. supply chain. Clearly, SCM must evolve to address the needs of the extended supply chain. In order to affect that evolution, however, we must first better understand the real 2. THE RESEARCH PROBLEM – DIGITAL problems that companies face in managing the extended QUALITY MODEL IN SCM supply chain. User demand for faster, more powerful technologies To meet the diverse and growing market demands for has shortened product lifecycles dramatically in the last quality and reliability, organizations realized that it is two decades, decreasing manufacturers’ opportunity to important to maintain and improve quality and reliability make profits on new products and forcing them to for product applications and usage environments. accommodate multiple product generations in their planning and production cycles. This acceleration has been accompanied by falling prices and smaller margins, 2.1 Supply Chain Situation especially on mature product categories where Supply chain has a big impact on success because differentiation lies in new features and marketing rather purchasing, holding and moving merchandize are very than in the core product functionality. costly. Collectively, U.S. companies spend a trillion As a result, companies have increasingly turned to dollars a year on their supply chains, just under 10% of contract manufacturers (CMs) as faster, more cost- the nation’s GDP [2]. About a third of this cost is for effective channels for procuring parts and producing holding inventory and the rest is for moving it around, products. These CMs can offer both economies of scale with a bit of change left over for administration. As large and access to less expensive labour pools as well as as these figures may seem, they used to be substantially specific manufacturing expertise. Moreover, by taking higher, totalling about 15% of GDP at the beginning of over production of a company’s core product lines, they the 1980s. allow OEMs to focus on differentiating their products The same percentage holds good for individual through the creation of new features and more effective companies, which spend an average of just under 10% of marketing campaigns. The result of this outsourcing has their gross income on supply chain functions. What is been the creation of broad virtual manufacturing striking about the figures for individual companies is the networks that often span the globe. tremendous advantage that some companies have over Despite their economic justification, these extended others in this regard. A recent survey of supply chain supply chains have created new barriers to productivity costs across a variety of industries yielded an average of because they increase the complexity of communication 9.8% of revenue devoted to supply chains, a perfect between participants. Manufacturers and their supply match to the overall value. But the survey also revealed chain partners typically handle communication across the that the top the 25% best performers had an average cost geographically dispersed supply chain using a variety of of just 4.2% of revenue. These companies spend less than means: e-mails for regular interaction, faxes for half as much on their supply chains as the competition, exchanges requiring documentation or real-time input and giving them a full five-point advantage in profits. phone calls for emergencies. Although there may be some Continuing surveys reveal that the gap is not closing, but sharing of information systems, the number of widening. According to a recent US Bancorp Piper participants and their varying business requirements Jaffray study, supply chain management is the number makes widespread installation and integration of a one strategic priority for manufacturing executives. But “single” information system impractical. the reality of SCM differs dramatically from its promise. As a result of the many parallel mechanisms that Consider some widely reported high-tech supply-chain make up supply chain communication, decision making, snafus from 2001 [3]. visibility and control are spread over many players, • Cisco Systems wrote off $2.5 billion in excess introducing more possible points of error, delay and inventory owing to poor management of its myriad problems, and effectively limiting companies’ ability to outsourced contractors. identify and respond efficiently to changes in the supply • Micron Technology wrote down $260 million of chain. The effect of this complex communication memory products inventory, representing 32 percent of scenario, in many cases, is distorted demand information, revenue. excess inventory levels, idle manufacturing capacity, SCM appears to be failing in its core mission of higher costs for fixing problems, and increasingly helping manufacturers control flows of goods and unhappy customers. information. The reason is that the typical supply chain, for which the majority of SCM solutions were developed, 2.2 New Quality Management Paradigm (Supply has evolved from an internally-owned function with a Chain Uncertainty and Risk) small number of players to a widely dispersed, largely outsourced function, with dozens or even hundreds of Quality has evolved through the following distinct participants across the globe. Communication and stages: visibility across this broad community have become the • Conformance focused - Quality is the degree to biggest manufacturing challenges for today’s companies. which a specific product conforms to a design or At the same time, markets have been driven by specification [4] evershrinking product lifecycles that leave little room for

110 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

• Market focused - To practice quality control is • Visibility of end-to-end process activities to develop, design, produce, and service a quality • Process components and functionality that are product, which is most economical, most useful, and exposed and self-describing always satisfactory to the consumer [5]. • Ability to integrate disparate information source • Excellence focused - Quality is commitment and application functionality into a process made real. It is not perfection. But, rather, the dedication • Information flow and event notification that can to perfection [6]. be automated and monitored throughout a process • Value focused - Quality is the relative • Workflow participation that makes the most of excellence of the composite of all product attributes in desktop productivity and communication tools fulfilling the needs and reasonable expectations of those • Service level agreements that can be specified, whom the product serves, as they perceive such monitored, and enforced for activities in a process fulfilment from time of offering throughout product life • Ability to add, remove, or reconfigure any [7]. process activity or component, without disrupting the Uncertainty and risk process Uncertainty exists because of globalization, • Processes that can be monitored in real time or technology, mergers, acquisitions, saturated markets, and near real time global competition. Uncertainty and risk arise from an • Process designs that can accommodate any inability to plan, execute, and ultimately control events. exception handling requirement Also, the likelihood and consequences that potential • Processes that can be easily replicated, extended, events may occur are now part of every management and scaled discussion in companies as well as government. Quality is critical to success, and in a competitive, Most senior management decision-making today is fast-moving world, it cannot be an afterthought. You filtered through a risk filter. Risk can be defined as a have to engineer quality into the design of products and situation or circumstance, which creates uncertainties processes; you have to track quality through procurement, about achieving program objectives [8]. Risk production, and delivery; and you have to constantly management represents an organized, systematic, improve quality to keep pace with ever growing customer decision-support process that identifies risk, assesses or expectations and to stay ahead of competitors. analyses risks, and effectively mitigates or eliminates In short, you need to take a comprehensive approach risks to achieving the program objectives [9]. Six sigma to quality management – one that integrates information methodology to define, measure, analyze, improve, and and processes across departments and corporate control (DMAIC) is fundamentally a risk management boundaries, and lets a range of employees and supply methodology. chain partners contribute to maintaining and enhancing quality. 2.3 Business Process Management Processes are important assets. They are a company’s core competencies and determine business performance. Competing in today’s business environment precipitates 3. QUALITY MODELS the need for successful integration and collaboration strategies. The reason why companies offer quality is to satisfy Business processes are dependant and ordered the customer. Companies that do not continually satisfy activities that result in predictable and repeatable their customers will lose those customers and as a result outcomes. Consisting of an organization’s operating they will achieve poor performance. Quality is seen as procedures, institutional working knowledge, and essential for strategic success (Garvin, 1988). Poor information resources, business processes are designed to quality will lead to losses of profitability and market satisfy defined business objectives in an efficient and share. In the next sections, we will present some widely timely manner. In an efficient environment, the functional adopted quality management models. components of a process can be readily identified, adapted, and deployed to address ever-changing corporate 3.1 ISO 9000:2000 requirements—a capability termed as business agility. Business Process Management (BPM) is the next Quality is defined in ISO 9000:2000 as: “the degree to innovation towards the agile enterprise by creating which a set of inherent characteristics fulfil automated, reliable, and rapid processes that are requirements.” There is no mention of product or service synchronized across functions, departments, and supply or entities so we can apply the definition to any set of chains. The benefits are lower costs and higher revenues. requirements - financial, environmental, safety, social, The vision behind the BPM is to enable business to economic and also functional, physical and human change processes ate the speed of market dynamics. BPM requirements, even maintenance. If we change the way is a closed loop model for engineering, executing, we think about quality, we see that it is not about monitoring, and controlling business processes. It bridges following procedures, inspection, rules and regulations. It the gap between business and IT. is about establishing the needs and expectations of those Automated business processes developed and we choose to serve, setting goals for satisfying these executed within such an environment are characterized by needs, devising a system of processes to fulfil these the following attributes:

111 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 goals, measuring performance and continually improving initiatives are not mutually exclusive and the objectives capability to satisfy the needs of all interested parties. in applying them are different. The ISO 9000 series is composed of: Management strategies, such as TQC, TQM, and Six • ISO 9000:2000 explains the concepts, the Sigma, are distinguished from each other by their principles and defines the terms. underlying rationale and framework. As far as the • ISO 9001:2000 specifies requirements for corporate framework of Six Sigma is concerned, it assessing the capability of organisations to meet customer embodies the five elements of top-level management and applicable regulatory requirements - it is not a design commitment, training schemes, project team activities, specification for a quality management system measurement system and stakeholder involvement [10]. • ISO 9004:2000 contains a guidance on At the core of the framework is a formalized performance improvement - it is not a guide to ISO 9000 improvement strategy with the following five steps: neither is it a design specification although it comes a lot define, measure, analyse, improve and control (DMAIC). closer than ISO 9001. ISO 9001 basically requires the organisation to: 3.3 Lean Manufacturing • Determine the needs and expectations of Lean evaluates the entire operation of a factory and customers and other interested parties; restructures the manufacturing method to reduce wasteful • Establish policies, objectives and the work activities like waiting, transportation, material hand-offs, environment necessary to motivate the organisation to inventory, and over-production. It reduces variation satisfy these needs; associated with manufacturing routings, material • Design, resource and manage a system of handling, storage, lack of communication, batch interconnected processes necessary to implement the production and so forth. policy and attain the objectives; While Lean emphasizes standardization and • Measure and analyse the adequacy, efficiency productivity, Six Sigma can be more effective at tackling and effectiveness of each process in fulfilling its purpose process noise and cost of poor quality. In a system that and objectives and; combines the two philosophies, Lean creates the standard • Pursue the continual improvement of the system and Six Sigma investigates and resolves any variation from an objective evaluation of its performance. from the standard. The focus is therefore on results and the processes In SCM, Lean concepts might include: that produce these results. This means that there needs to • The preparation of the supply base material be a linkage between the needs of the interested parties, delivery systems, including: lot size identification and the organisation's objectives, the processes for achieving compliance, and standard packaging development or these objectives and roles being produced. enforcement. The old methods of documenting what you do and • The implementation of increased shipment doing what you document will not cause the right things frequencies where: Smaller lot sizes are driven by closer to happen. The old methods of auditing for conformity to control of manufacturing space and asset utilization, the requirements or procedures will not verify that processes need for formal frequency planning and cost evaluation are being managed effectively and therefore both have to becomes clear, and milk-run consolidation, cross-dock, change. and other logistic issues are addressed. By looking at ISO 9000 as a framework upon which can be built a successful organisation, rather than as a 3.4 National Quality Awards narrow set of a minimum requirements, significant benefits can be gained. There are real benefits from The national quality awards models such as the managing organisations as a set of interconnected Malcolm Baldrige National Quality Award (MBNQA), processes focused on achieving objectives that had been the European Quality Award, and the Deming Prize derived from an understanding of the needs of customers provide a set of excellent similar criteria for helping and other interested parties. companies to understand performance excellence in operations. Self-Assessment is heavily diagnostic in 3.2 Six Sigma nature with most criteria that guide companies towards excellence. Six Sigma (6σ), as a concept, is a business-driven, multi-faceted approach to process improvement, reduced costs, and increased profits. With a fundamental principle to improve customer satisfaction by reducing defects, its 4. SCM DIGITAL QUALITY MODEL ultimate performance target is virtually defect-free processes and products. Six Sigma is a revolutionary business process geared toward dramatically reducing Companies seek to differentiate themselves from their organizational inefficiencies that translates into bottom- competitors not only by the quality of their products and line profitability. services, but increasingly by offering additional services Six Sigma, at its basic level, is attempting to improve to actively winover customers. We see ISO 9001 as an both effectiveness and efficiency at the same time. Six international consensus of good management practices Sigma is a measure of customer satisfaction that is near and set about making it the starting framework for how perfection. Six Sigma is needed regardless of whether a the organization would operate. company is compliant with the ISO 9000 series. The two

112 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Having in mind all previously analyzed business and models), and the functional domain (information market drivers, and the lack of well-defined supply technology infrastructure, open software standards, network excellence model and the methodology for modern object-oriented development methods, and architecting the model, we started to work on this subject. cutting-edge software tools). This work is the subject of The idea was to create methodology which unifies the undergoing research efforts [11]. business domain (processes, people, existing supply chain Figure 1 shows the master scheme for achieving reference models, best practices, and quality management supply chain management excellence.

Figure 1. Supply Network Excellence Framework

113 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

For every digital product or service model, supply network system needs to be developed. Within this system, quality requirements must be fulfilled by all supply network members. Effective information management about product or services, and technological resources are the key elements of digital factory which can be defined in three levels [12]: 1. Defining of the total supply network model (product, service, process, execution) that represents the prototype model for planning. 2. At the second level, static model is transformed into dynamic using the information technologies and simulation tools and techniques. The result is virtual supply chain which can be optimized by different criteria, mostly by unit costs. Virtual configuration can be changed and run easily. Figure 3. Different views of one system in one model 3. The third level is fusion of virtual and real model, and the real supply network. After the supply That’s why we propose the Rational Unified Process network realisation, the data are embedded into the (RUP) as a basic methodology for business modelling simulation model. and later for application development [14]. RUP support The starting point is supply network modelling. This object-oriented development and it allows to be step is very important because we need to know what configured for particular use. RUP enables quality processes exist, the way they are performed, actors, and management and risk management. Quality is built into the information flows. Every error in this step will result this process in all activities and encompasses all in huge costs incurred in later stages. That’s why we participants, with use of appropriate metrics and criteria. propose model driven development based on MDA On the other hand UML (Unified Modelling Language) (Model Driven Development) [13]. The Model Driven provides standard notation and with nine different types Architecture (MDA) is a framework for software of diagrams offers rich modelling (Figure 3). development defined by the Object Management Group Another business modelling methodology is (OMG). Key to MDA is the importance of models in the UN/CEFACT Modelling Methodology (UMM). UMM is development process (Figure 2). The first model that an incremental business process and information model MDA defines is a model with a high level of abstraction construction methodology that provides levels of that is independent of any implementation technology. specification granularity suitable for communicating the This is called a Platform Independent Model (PIM). In model to business practitioners, business application the next step, the PIM is transformed into one or more integrators, and network application solution providers Platform Specific Models (PSMs). A PSM is tailored to [15]. The UMM provides the conceptual framework to specify your system in terms of the implementation communicate common concepts. It also based on RUP. constructs that are available in one specific The other important aspect of business modelling is implementation technology. In terms of improvements in use of business standards. They can be functionally or development process MDA foster productivity, enables industry specific, with the goal to facilitate and simplify portability and interoperability, easier maintenance, and business collaboration and information exchange. When consistent documentation. it comes to supply chain and e-business, there are already some widely accepted standards such as: SCOR (Supply Chain Operations Reference) Model, Odette SCMo (Supply Chain Monitoring), ebXML, and Open EDI Reference Model. For our work we based our modelling on the SCOR model because process reference models integrate the well-known concepts of business process reengineering, benchmarking, and process measurement into a cross-functional framework. Some recommendations from other reference models are also included. The result of the fist phase modelling is the SCM Model presented in UML notation which presents comprehensive business process and business Figure 2. Business and software models information model. It also helps discover and define a set of reusable process and information descriptions - patterns. Patterns help enforce consistent, reproducible results from across business domains. Figure 4 shows sequence diagram of inventory management process, and Figure 5 shows the segment of database scheme – quality management – development process.

114 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

: order list : order list : : work table : order items creation_us creation_bs warehouse_product_model 1: start program() 2: load row() 3: check model type(model type)

{model type="FQS"} 4: check level(level < ROP)

5: insert row() {vrsta modela="FPS"}

6: inspect order time()

7: insert row() 8: check special orders(order type)

9: special orders()

10: insert row()

11: check urgency orders(express orders)

12: express orders()

13: insert row()

Figure 4. Inventory Management Process

T_Validation T_production program val_number : INT <> T_Dev elopment (from S_0) product_ID : INT product_ID : INT partner_ID : INT partner_ID : INT process_ID : INTEGER product_ID : INT v al_description : CHAR(1) 1 0..* v al_date : SMALLINT <> PK_T_projektov anje108() <> PK_T_proizv odni program35() <> FK_T_projektov anje148() <> FK_T_proizv odni program49() <> PK_T_v alidacije115() <> FK_T_proizv odni program50() <> TC_T_v alidacije1527() 0..1 0..* <> FK_T_v alidacije153() <> <> 1

1 T_Product (from S_0) product_ID : INT T_Verif ic ation T_production_process T_Process <> <> product_name : CHAR(30) <> v er_number : INT process_ID : INTEGER process_ID : INTEGER product_ty pe : INT product_ID : INT product_ID : INT process_name : CHAR(1) tariff _group : INT v er_description : CHAR(1) 1 0..1 0..*1 weight : INT 1 0..* v er_date : DATETIME <> PK_T_proizv od_proces83() <> PK_T_proces82() volume : INT <> FK_T_proizv od_proces111() <> TC_T_proces963() material_ty pe : INT <> PK_T_v erif ikacija114() <> FK_T_proizv od_proces112() ABC group : CHAR(1) <> TC_T_verif ikacija1526() 1 1 <> FK_T_v erif ikacije154() <> PK_T_proizv od0() <> <> 0..1 0..*

T_Procedures T_Project T_Activities proc_number : CHAR(1) project_number : INT activity_ID : INT proc_description : CHAR(1) project_name : CHAR(1) <> activ ity _name : CHAR(1) proc_date : DATETIME author : INT product_ID : INT process_ID : INTEGER project_date : DATETIME process_ID : INTEGER product_ID : INT 1 0..* project_number : INT <> PK_T_procedure109() process_ID : INTEGER <> FK_T_procedure149() <> PK_T_aktiv nosti111() <> PK_T_projekat110() <> TC_T_aktiv nosti1450() <> TC_T_projekat1433() <> FK_T_aktiv nosti151() <> FK_T_projekat150()

Figure 5. Quality Management – Development Process

115 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

The next step is incorporation of quality management information technology and computer simulation system, based on ISO 9000 series of standards, and achievements is imposed. We purpose supply chain extended with some recommendations from Six Sigma simulation based on the developed supply chain model and CMMI (Capability Maturity Model Integration). [16]. Thereby, the framework for optimization, scenario Also, maintaining the knowledge and best practices analysis, risk management, project planning, and virtual database helps improve modelling quality and mitigate quality can be established. repeating of errors. This synergy results in SCM Quality • Adaptive Supply Network - An adaptive supply Model. Thanking to the use of unified modelling chain has several dimensions. It is essential to create methodology and notation, using open standards like good plans and good forecasts and well, have a good XMI (eXtensible Model Interchange) and XML overview of what we think what is happening in the (eXtensible Markup Language), this model facilitates the future. We need to plan, execute, sense, and to respond specification of reusable/reproducible process models, in [17]. And finally if we have done all this we would like to objects and interface-specific object behaviour improve continuously. We would like to learn from what descriptions that are technology and protocol insensitive. we did in the past. Software agent system based on Using the realization techniques and with the support distributed software platform is the enabler of adaptive of specific compilers, this model can derive database framework. schemes, classes, components in the target software platform. The result is SCM Digital Quality Model. It is the central point in SCM software development, because it provides the link with the business domain. This way, 5. CONCLUSION AND FUTURE RESEARCH any changes in business domain can be reflected in functional domain, thus ensuring consistent software The surge of competitive factors in the past few years design. Software patterns (creational, structural and has forced companies to constantly rethink and improve behavioural) also can be engineered from this model. internal as well as external business processes. Innovative Using the enabling technologies (Web Services, products which are quickly available and low in cost, Agent Software, and FRID-Radio Frequency while meeting the highest standards for quality and Identification) which foster supply network, the final customer service, are essential for success. The software systems can be developed. We classify them in knowledge and development of internal and external four system categories: company business processes in connection with • Business Process Management - The next era of organizational changes represents a big challenge. Both computing will be characterized by the detachment of customers and suppliers have to be involved, and the information from applications, leading to a widely company’s own resources have to be optimally planned distributed Service Oriented Architecture. Just as and utilized. standards-based Web servers and browsers facilitated the Companies that want to understand how their communication and distribution of information between organization functions have to see this as a complex and people, BPM tools that employ standards-based XML dynamic group of processes. Quality management is the and Web Services technologies will facilitate the wide- ideal tool to guide the company in all areas and tasks and scale proliferation of automated and distributed business to increase efficiency. International quality norms and processes. A defining characteristic of BPM technology models have recognized this and treat the management of is the elevation of design and development functions business processes as a basic prerequisite for the success from the program layer to the information (document) of a company. The primary requirement of these and transport (messaging) layer. Workflow processes, worldwide standards is an effective and consistent integration scenarios, or trading partner interactions modelling of business processes. consist of an orchestrated flow of messages that are The new shift in the global business arena has routed, transformed, and processed according to message redefined the way companies work together, raising content, formatting requirements, and business rules. customer expectations and imposing quality requirements • Business Intelligence - BI goes beyond static far exceeding existing standards. This paper discusses the data snapshots to enable users to identify and analyze new business paradigm, analyses existing quality ongoing business trends and patterns. By providing wider management standards, and propose the unified visibility to plans and supporting data, BI tools increase framework for achieving supply chain management the return on existing SCM applications because they excellence through synergy of modelling techniques, help companies understand where and how they deviate supply chain reference models, quality management from their plan objectives. In addition, they provide models, and the latest software technology for building shared data availability that encourages a global flexible supply network capable to support many-to-many perspective on business performance. As a result, people real-time collaboration. are more likely to make decisions based on their global The future work will be carried out to develop impact thus increasing quality. methodology, metamodel, and finally SCM digital quality • Simulation - Considering the SCM complexity, model based on presented unified framework. The final the question arises how to design and plan SCM that can goal is to achieve requirements quality, performance provide supply chain profitability? The changes in one quality, and customer quality. This level of quality can be supply chain component usually create waves of ultimately integrated with the environment management, influence that propagate throughout supply chain. It is work safety and health protection into the quality of life. obvious that the necessity of application of modern

116 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

REFERENCES

/1/ H. Lee, Aligning Supply Chain Strategies with Product Uncertainties, California Management Review Vol. 44, No. 3, Spring 2002. /2/ David A. Taylor, Supply Chains: A Manager's Guide, Addison-Wesley, 2004. /3/ From Planning to Control: Improving The High-Tech Supply Chain, Valdero Corporation, 2002. /4/ Gilmore, H. L., Product Conformance Costs, Quality Progress, June, 1974, p. 16. /5/ Ishikawa, K. What is Total Quality Control? , Englewood Cliffs, NJ: Prentice Hall, 1985, p. 44. /6/ Copy in advertisement for Shearson Lehman Brothers, Wall Street Journal, November, 1987, p. 26. /7/ L. J. Utzig, Quality Reputation - A Precious Asset, 34th Annual Quality Congress Transactions, Milwaukee, WI: ASQC, 1980. /8/ FAA Programmatic Risk Management, 2002, p. 6. /9/ COSO, Enterprise Risk Management Framework, web, 2003. /10/ Sung H. Park, Six Sigma for Quality and Productivity Promotion, Asian Productivity Organization, 2003. /11/ N. Stefanovic, Development of Digital Quality Model in Supply Chains, PhD Thesis, Mechanical Engineering Faculty, University of Belgrade, Serbia and Montenegro, 2005. (ongoing work). /12/ V. Majstorovic, Digital Factory – Fiction, Reality or Future, 30th Jupiter Conference, Belgrade, Serbia and Montenegro, 2004. /13/ Frankel, David. Model Driven Architecture: Applying MDA to Enterprise Computing. New York: John Wiley & Sons, 2003. /14/ Rational Unified Process, Rational Software Corporation, 2003. /15/ UMM User Guide, UN/CEFACT, 2003. /16/ D. Stefanovic, N. Stefanovic, V. Majstorovic, Methodology for Process Integration in Supply Networks, 38th CIRP ISMS, Brazil, 2005. /17/ N. Radjou, Migrating To An Adaptive Supply Network, Forrester Research, 2003.

117

Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

ISO MANAGEMENT SYSTEMS STANDARDS AND SOCIAL RESPONSIBILITY CONNECTION: (NOT QUITE) JOINED-UP OPINIONS OF ISO’S STAKEHOLDERS

Pavel Castka 1), Michaela A. Balzarova 2) 1) University of Canterbury, Christchurch, New Zealand, 2) Brno University of Technology, Brno, the Czech Republic

Summary: It has been suggested in the literature that if TQM is to remain a viable concept into the 21st century it must embed more deeply and firmly the issue of virtue within its domain. Indeed, recent research into quality-social responsibility link and efforts by International Organization for Standardization (ISO) to develop a standard in social responsibility confirms this trend. The purpose of this paper is to present and analyze the opinions of major stakeholder groups on standardization of social responsibility by International Organization for Standardization (ISO). We have collected the data at International Conference on social responsibility held in June 2004 in Stockholm, where 355 delegates from 66 countries, invited by ISO, have participated. The data consists of key note addresses, stakeholder presentations and panel discussion, which were recorded and transcribed verbatim. To analyze the data, we have used NVivo - a qualitative data analysis software. As a result four high level topics have emerged: standardization (is SR standard desirable?), guidance document (should ISO proceed with the guidance document instead?), definition (how do we define SR and is its scope?) and impeders (what are the impeders in the pursuit of the SR agenda?). We closely discuss these areas and reveal often contradictory issues that were pointed out by different stakeholder groups. Keywords: ISO management systems, corporate social responsibility, standardization

INTRODUCTION opinions of major stakeholder groups on standardization of social responsibility by International Organization for Over the last few years, academics, governments and Standardization (ISO). practitioners alike have shown a growing interest in corporate social responsibility (CSR) and CSR has started to infiltrate in the field of quality management as well. A ISO’S INVOLVEMENT IN THE FIELD OF number of studies has looked at integration of CSR into 1 management systems (Castka et al., 2004a,b; SOCIAL RESPONSIBILITY Karapetrovic and Jonker, 2004) and similarly a link between quality and ethics has been investigated more In ISO Bulletin (2002), ISO announced its intention to closely (Grayson, 2003; Hodkinson, 2003). Following the grasp this area with the further possibility to develop a discussion of Park-Dahlgaard (1999), Ahmed and standard. Yet from the very beginning, ISO also Machold (2004) reflect on the evolution of TQM and questioned whether the standard is actually the right assert: mechanism to advance CSR (ISO Bulletin, 2002). In the “Park-Dahlgaard (1999) indicates the stages of meanwhile a number of national standardization bodies evolution of the TQM, highlighting the widening and (for instance, France, Australia, Israel, Japan, Mexico, deepening of the scope. The four major stages of the UK and many others) progressed with national standards quality movement have been described as inspection, or guidelines for SR (Castka et al., 2004a; Göbbels, statistical quality control, quality assurance, and total 2003) and other standards outside ISO were developed quality management. Their respective primary concerns (AA 1000, 1999; SA 8000, 1999; for comparison see are detection, control, coordination, and strategic Göbbels and Jonker, 2003 and Rocha and Karapetrovic, impact. We suggest that if TQM is to remain a viable 2003). concept into the 21st century then it must embed more The standardization process at ISO goes through a deeply and firmly the issue of virtue within its domain of number of specific phases (Tamm Hallstrom, 2000). In content.” terms of SR, ISO formed firstly ISO COPOLCO (ISO The view of Ahmed and Machold (2004) can also by Committee on consumer policy) and later ISO Advisory supported by recent initiatives within International Group on Social Responsibility (AG SR) to deal with Organization for Standardization (ISO). In 2004, ISO initial stages of SR standardization. Both ISO COPOLCO Technical Management Board has prepared a formal and ISO AG SR produced reports that provided a basis proposal that ISO undertake the preparation of an for wider discussion (see Table 1). International Standard giving guidance on social responsibility. ISO/TMB/WG SR (2005) in its draft project plan aims to publish SR international standard in 2008. In this paper we closely look at developments 1 ISO used corporate social responsibility (CSR) or corporate within ISO and we aim to present and analyze the responsibility (CR) in the past. The latest documents use the term social responsibility (SR). 119 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Year Committee/AG/ Report Refere 2. ISO recognizes that it does not have the authority or Event nce legitimacy to set social obligations or expectations 2002 ISO COPOLCO The Desirability ISO which are properly defined by governments and (ISO Committee and Feasibility of COPO intergovernmental organizations. on consumer ISO Corporate LCO 3. ISO recognizes the difference between on the one policy) Social (2002) hand, instruments adopted by authoritative global Responsibility Standards inter-governmental organizations (such as the United Nations Universal Declaration on Human Rights, 2003 ISO Advisory Working Report ISO Group on Social on Social AG SR international labour conventions and other Responsibility Responsibility (2004a, instruments adopted by the ILO and relevant UN b) Conventions) and on the other hand, private voluntary 2004 ISO Press release; Ref ISO initiatives that may or may not reflect the universal International 921; 24 June 2004 (2004) principles contained in the above instruments. Conference on 4. ISO narrows the scope of the subject so as to avoid Social addressing issues that can only be resolved through Responsibility political processes. 2004 Technical New work item ISO/T 5. ISO recognizes through a formal communication the Management proposal – Social MB ILO's unique mandate as the organization that defines, Board prepares responsibility (2004) on a tripartite basis, international norms with respect formal proposal to a broad range of social issues. for the 6. ISO recognizes that, due to the complexity and fast- preparation of the International evolving nature of the subject, it is not feasible to Standard harmonize substantive social responsibility Table 1 ISO’s involvement in the field of social commitments. responsibility 7. ISO reviews its processes and where necessary makes adjustments to ensure meaningful participation by a Let us firstly summarize the events. In 2002, fuller range of interested parties. ISO/COPOLCO (ISO Committee on consumer policy) These recommendations were further discussed at announced that a trio of ISO management standards - ISO ISO conference on social responsibility, which is the 9001, ISO 14001 and ISO corporate social responsibility focus of this paper. Our aim is to present and analyze management systems standards (CSR MSS) – would opinions of participants of this conference. support business efforts to show that an organisation cares about quality, environment and the social effects of the production or activity (ISO COPOLCO, 2002). In their report, the committee has, amongst other issues, RESEARCH FOCUS: DISCUSSIONS AT ISO recommended that (ISO COPOLCO, 2002): SR CONFERENCE “ISO CR MSSs would constitute an internationally agreed-upon framework for operationalization of ISO Conference on Social Responsibility took place corporate responsibility commitments, capable of in Stockholm (June 2004). Participation in the conference producing verifiable, measurable outputs. The ISO CR was open to ISO member body delegations and invited MSSs would build on the intellectual and practical international groups and organisations. infrastructure of ISO 9000 quality MSSs and ISO 14000 The conference was hosted by the Swedish Standards MSSs, and the momentum associated with close to one- Institute (SIS), and drew together 355 participants from half million firms certified as compliant with these 66 countries, including 33 developing countries, standards. As with ISO 9000 and ISO 14001, firms could representing major stakeholder groups: national standards self-declare compliance with the proposed ISO CR MSSs institutes, business, government, labour, consumers, or could seek certificates from authorized third parties. It international and nongovernmental organizations. The should be emphasized, however, that ISO CR MSSs would recommendations of the AG SR (ISO AG SR, 2004b) be insufficient by themselves to assure that a firm has made up the backbone of the discussions at the developed and implemented an effective CR approach. conference. The conference consisted of plenary sessions, Thus, ISO CR MSSs would be one piece – albeit a stakeholder presentations and panel discussions, and fundamental building block – of effective CR parallel breakout sessions. We have recorded all approaches.” presentations and discussions, which were then Whereas ISO COPOLCO’s report linked SR transcribed verbatim. We used NVivo, a software for initiatives to ISO 9000 and ISO 14000 and “management qualitative data analysis to code and analyse the data. As by objectives” approach, the report of ISO AG SR moves we coded the data, we have identified a set of prevailing the focus on legitimacy of such initiative. The report by high level topics (see Table 2). Table 2 presents overview ISO AG SR concluded that ISO should only proceed if of these topics and clarifies the key question for each of (ISO AG SR, 2004b): them. This is followed by the results for each topic, 1. ISO recognizes that social responsibility involves a which are further discussed in the next section of this number of subjects and issues that are qualitatively paper. different from the subjects and issues that have been already dealt with by ISO. 120 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

Table 2 Results Topic Key question Results Standardization Is the SR standard desirable? ƒ Mostly supported by delegates from developing countries ƒ Opposed by multi-nationals ƒ Voting during the plenary session: roughly half of the delegates in favour, half against Guidance document Should ISO proceed with the ƒ Proposed by the Advisory Group on SR guidance document? ƒ Supported by delegates representing NGOs and some multinationals (those not supporting the idea of standardization) ƒ Opposed by other multinationals (those in favour of standardization) ƒ Opposed by the delegates from developing countries Definition How do we define SR and ƒ Multinational industries: we need a level of what is its scope? common understanding first before we proceed any further ƒ It is not clear how far does SR go upstream ƒ SR is extremely broad – different stakeholder groups put different emphasis on particular aspects of SR ƒ Organisations do not define their own SR, society defines responsibilities for them ƒ Most multinationals create their own code of conducts that are reinforced through the supply- chain Impeders What are the impeders in the ƒ SR claims to go ‘beyond legal compliance’ yet pursuit of the SR agenda? legislation in not reinforced in many countries ƒ Similarly, labour standards are not implemented by local governments ƒ How can we assess SR compliance? SR can be hijacked from public relations purposes or by activist groups

Discussion asserted that SR is a global issue and ISO must take into The results reveal that the standardization of the SR consideration that 75% of ISO’s members are from agenda is far from a global multi-stakeholder consensus. developing countries. One of the delegates explained why Indeed, to find a consensus globally and with developing countries support ISO: stakeholders with such a variety of backgrounds (national “…one of the things we are concerned with is that standards institutes, business, government, labour, many CSR initiatives are imposing upon developing consumers, international and nongovernmental countries some constrains for our market or for organisations) will require more than a conference. A marketing our products….it would be better to have just brief overview in Table 2 highlights two broad issues. one leader then having hundreds of different schemes.” Firstly, there are issues that delegates could not agree The AG Group on Social responsibility recommended upon (standardization, guidance document) and on the that ISO should proceed with guidance not a specification other hand, issues that are still unclear (definition, document (see the introductory section of this paper). impeders). One of the AG SR members explained: “The advisory group recommends guidance not a Disagreement – standardization versus guidance specification document against which the conformity can document be assessed……. The background, the discussion of this Many delegates expressed a strong believe that ISO is topic was that classical management system approach capable of producing standards that represent the global will not be promoted… enhancing credibility - that is the consensus amongst wide variety of stakeholders area where ISO can add value…. ISO guidance document (“building consensus” was the catch phrase as ISO uses a must be more than addressing management systems consensus based dialogue process as a mean to standard, it should give the practical guidance on developing standards). The reputation of ISO as a diversity of issues and how to handle them.” standards body was echoed by many of the delegates One of the delegates from the NGO community stated representing developing countries. These delegates often that for NGOs, a management system framework is not stressed that “there is a need to align codes and agreeable: “There is a very strong concern within the initiatives” and that this could bring “the acceptance of NGO community that ISO document will become the more global approach by national governments and alibi document to use the standard for advertising stakeholders, who can chose ISO standards instead of purposes and that is the reason why we want to have a developing national of regional”. These delegates also standard that contains ethical requirements and not the

121 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 management system standard”. Similarly, delegates from are fundamental elements of social responsibility. Is one multinationals opposed the standard. As one of the of these more important to standardization? Finally, delegates put it “…of course we want to doing the there is an issue about consistency versus creativity. business…but we want to do it in a responsible way…. There are elements of social responsibility that take standards have the role to play but too often they are too sustained effort and consistent action process. There are general and it is too often too easy for a supplier to have a other that are innovative, creative, opportunistic.” nice certificate… there is a lot of our suppliers being ISO One of the most repeated issues in relation to 9001 and ISO 14001 approved but still not performing to definition of SR, was the role of the regulatory system. [our] standards, I can tell.” Other business community One of the delegates stated: delegates argued whether there is a real value of SR “…the assumption is that exists the robust regulatory standard and not for a very limited number of consultants systems around social issues. In certain countries with who make business out of this and many of these less robust regulatory system they might need some supported the guidance document. Some of these guidance in this area. If your inclusion factor, the criteria delegates stressed that certification would not bring any you use for what is in and what is out, it based on what is value for the company and its suppliers and that they regulated, than by which country you talked about would only put millions into certification. Other delegate regulation - because if it is out of the local regulation it is pointed out at the possible misuse of standardization: out of the standard. I do not know the way round it, I am “…ISO 9001 and ISO 14001 are voluntary, but as soon just pointing out. The guidance without requirements is a as some activist groups start a mandate that you can sell substitute for the regulatory system.” and must have a SR standard on this or that market; it Indeed, delegates from most of the stakeholders group will become quasi-mandatory to do a business. And we strongly argued that the problem is that local raise the bar, we raise the cost of our products.” The governments are failing to implement their own law and argument of cost was refused by other business delegate: what is written in the labour standards. These seemed to “I know that there might be a cost that customers will be the most mentioned impeder to the progress. Further to have to pay but they are also paying $200mil fines for this, as we have already mentioned in the previous doing bad marketing. So we need to discuss what is section, there is questions of verification of ethical pay and what is SR for organisations not just “compliance”. Here the delegates were divided amongst companies. And also take into consideration countries those who were in favour of the standard (here the where governments don’t have the same strengths and argument was “if you develop guidance you cannot report same resources for doing this. We need to be fair as to it”) and those who were not. organisations. We need to help each other to move this forward because it is our responsibility.” However developing countries delegates were almost unanimously in support of the development of the CONCLUSION standard. One of the delegates stressed that “standardisation in this area could also help to promote In this paper, we provided an insight into the process greater efficiency” and that “ISO CSR standard is needed of SR standardization – a new challenge for ISO for to enforce the implementation of the regulations and coming years. We have presented key topics and legislation related to human rights, labour rules and the arguments and perhaps raised more questions than environment.” provided answers on them. Yet this was an outcome of recent discussions within the ISO community. What Confusion – definition of SR and other impeders seems to be eminent is the fundamental difference in The topic of how SR is defined and what is the actual views on what (C)SR really means. In other words, is the scope of SR seemed eminent for all keynote speakers and “invisible hand” linked to social responsibility or is it not also for other delegates at parallel sessions - many have and hence do we need “something” else? As the addressed this issue and pointed at inconsistencies in Economist (2005) put it, those thinking in Adam Smith’s defining and understanding SR. This was clearly seen as terms believe in free market, they should not oppose an obstacle for future discussions and SR standard market restrictions of any kind. Yet if we look at development. One of the AG SR members explained the arguments of delegates from developing countries, they view of the advisory group and stated that “CSR is mostly are in favour for SR standard for these reasons. extremely broad…it touches a wide range of people in a ISO already prepared a formal proposal for the lot of organisations and that is maybe one of the reasons preparation of the international standard (ISO/TMB, why the AG moved from CSR into SR - because it 2004) – to be lead by Brazil and Sweden. As the result of impacts on every one of us. You need a lot of expertise, the conference suggests (ISO 2004), it is likely that the different types of expertise just to get your mind standard is to be guidance based – possibly a new trend in around….” Other AG member cited the traditional standardization (for instance ISO 10000 development – definition that “CSR is a voluntary …… in addition to Dee et al., 2004). meet the legal requirements” yet at the same time pointed It is clear the SR agenda presents a challenging area at a bias in understanding SR. As other delegate put it: for the future research - many of those could be “…it is still quite unclear what the scope is. Is it synthesized from this paper. For researchers/practitioners about core activities or non-core activities? from the quality field, the discussion and (non)acceptance Standardization is perhaps related to core activities. Is it of ISO 9000 management system and continuing about managing risk or identifying opportunities? Both improvement approach might provide inspiration for 122 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 revision of quality field as is and its future direction. We accountability standards”, Managerial Auditing Journal, Vol.18 would like to conclude with two opinions raised during No.1, pp.53-58. the conference that were particularly inspirational and /7/ Grayson, D. (2003), “The CSR brand”, Quality World, could provide future developments in this area. January 2003, pp.10-13. /8/ Hodkinson, P. (2003), “Check your response”, Quality “The other fundamental issue is if we talk about the World, January 2003, pp.28-30. international standard, we talk about the international /9/ ISO AG SR (2004a), Working Report on Social generic standard; there is a difference between, sort of, Responsibility, ISO Advisory Group on Social Responsibility. principles and systems. If you set the lowest common /10/ ISO AG SR (2004b), Recommendations to the ISO denominator for systems and processes that you are using Technical Management Board, Document: ISO/TMB AG CSR - that is also a different issue the setting the lowest N32, International Organization for Standardization. common denominator of the principles. Is there the /11/ ISO Bulletin (2002), A daunting new challenge – are lowest common denominator? One of the issues that standards the right mechanism to advance corporate social numbers of us [in the advisory group] were thinking responsibility, ISO Bulletin, July 2002, International Organization for Standardization. about is the way the international standards get /12/ ISO COPOLCO (2002), The Desirability and implemented. Canada will adopt the minimum Feasibility of ISO Corporate Social Responsibility Standards, international standard but Canada can change the Final report by the Consumer Protection in the Global Market international standard if they have their own national Working Group of the ISO Consumer Policy Committee stakeholder to assess and say OK this is an international (COPOLCO), May 2002. standard and this is how we gonna interpret it in Canada. /13/ ISO (2004), “ISO weights results of conference on This is perhaps one way from moving from just social responsibility”, Press release Ref.: 921, 24 June 2004, international standard.” International Organization for Standardization, available on: “… there is some reason for companies to object www.iso.ch. /14/ ISO/TMB (2004), New work item proposal – Social standardization, if they think that a standard is something responsibility, International Organization for Standardization, imposing roles and imposing requirements, which is available on: www.iso.ch/sr. something people working in business ethics and CSR /15/ Karapetrovic, S. and Jonker, J. (2004), “Integration of trying to do. To explain I will give an example which we standardized management systems: searching for a recipe and use to teach business ethics. There is one definition of ingredients”, Total Quality Management, Vol.14 No.4, pp.451- business ethics that I like very much, and it is ‘business 459. ethics is a taught process we use to translate principles /16/ Park-Dahlgaard, S.M. (1999), “The evolution patterns into action’. So business ethics is not about teaching of quality management: some reflections on the quality people about what you should and should not be doing. movement, Total Quality Management, Vol.10 No. 4/5, pp. 473-481. Which is I think the same fear of the companies in /17/ Rocha, M. and Karapetrovic, S. (2003), “Merging relation to CSR standard. It is not about what companies Corporate Social Responsibility with an Integrated Management should or should not be doing. Business ethics school is System Framework”, Research Conference Managing on the helping people to decide about what is right and what is Edge, Nijmegen, the Netherlands, September 25 and 26, CD- wrong. The same the standard should do to help people ROM. to decide, to define what are their own social /18/ Temm Hallstrom,K. (2000), “Organizing the Process of responsibilities. So it is a tool to help them to take on the Standardization”, in: A World of Standards, Brunsson et al responsibilities.” (Eds), pp. 85-99, Oxford: Oxford University Press. /19/ The Economist (2005), “Profit and the public good”, The Economist, Jan 22 2005, Vol.374, Is. 8410, p.15.

REFERENCES:

/1/ Ahmend, P.K. and Machold, S. (2004), “The Quality and Ethics Connection: Toward Virtuous Organizations”, Total Quality Management, Vol.15 No.4, pp. 527-545. /2/ Castka, P.; Bamber, C. and Sharp, J. (2004a), Implementing effective Corporate Social Responsibility and Corporate Governance – A Framework, London: British Standards Institution. /3/ Castka, P.; Bamber, C.; Bamber, D.; Sharp, J. (2004b), “Integrating Corporate Social Responsibility (CSR) into ISO management systems – in search for a feasible CSR management system framework”, The TQM Magazine, Vol.16 No.3, pp.216-224. /4/ Dee, B.; Karapetrovic, S. and Webb, K. (2004), “As Easy as 10001,2,3”, Quality Progress, June 2004, pp.41-45. /5/ Göbbels, M. (2003), “Standardizability of Corporate Social Responsibility”, Managing on the Edge Shift in the Relation between Responsibility, Governance and Sustainability, Nijmegen, the Netherlands, September 2003. /6/ Göbbels, M.; Jonker, J. (2003), “AA1000 and SA8000 compared: a systematic comparison of contemporary

123

Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

RESEARCH INTO THE POSSIBILITIES OF THE APPLICATION OF THE FUZZY LOGIC TO THE DEVELOPMENT OF THE QUALITY EVALUATION MODEL

Marko Mirkovich, M.Sc.1), Prof. Dr. Dragan Radojevich2) , Prof. Dr. Vidosav D. Majstorovich3), Prof. Dr. Janko Hodolic4) 1 Bauxite mines,s.c. Quality & Information Centre , Niksic 2 Mihajlo Pupin Institute ,Belgrade 3 Mechanical Engineering Faculty , Laboratory for Production Metrology and TQM , Belgrade 4 FTS, Production Mechanical Engineering Institute , Novi Sad

Summary:Briefly, this work presents the results of the research into the possibilities of the application of the fuzzy logic to the field of quality, that is the evaluation of the same. The development of the methods (models) of the quality assessment / evaluation whose basic feature is the increasing objectivity of the same name procedures is what many researchers are interested in. The application of the fuzzy logic and fuzzy sets in that sense is one of the possible ways of reaching the set goal and the examples of the experiment that was carried out prove that was carried out prove that. KeyWords: Quality, evaluation, fuzzy logic.

1.INTRODUCTION theory, (ii) main educational profile-orientation and (iii) practical experience, of the researcher. The word quality derives from a Latin word qualitas Considering the said differences it is possible to meaning: feature, characteristic, and qualis meaning kind, identify in this field two main research frameworks: type, quality. Quality is, therefore, the general term and - formulation of the general theory of quality (various its use is possible and clear if it is " joined / ascribed" to models of quality, etc.) and the concrete entity. In this context "the entity is - development of the quality within the current everything that can be described and discussed developed fields (organization, economy, docimology,..). individually, such as : The main feature of the first approach is that it is -activity or process, based on the modern concept of quality(quality in the true -product, sense of the word)while the second approach is in essence -organization, system, person, abstract term or any traditional- specialty. In the traditional approach the term combination of the already mentioned." and the meaning of quality is strictly “imprisoned” by the Many definitions of the quality in the literature are the subject and the domain of the source area. The effort to best evidence of attractiveness as the phenomenon of the prove all-inclusive and the possibility of the source area research work. The current definitions describe the to integrate the quality into its concept /plan is very author’s approach to the quality research and without obvious in this approach. exception they “confirm” that the term “quality” is Contrary to this approach ,the modern approach ,that inclusive in all areas of human work, especially in the accepts in the first place the results of the researched exchange of goods and services. So, it is possible to evolution of the term and meaning of quality , makes the conclude that the quality is the main relation between the pragmatic concept and the widely acceptable research creator and the consumer of the same. This relation is the plan (platform) of the quality phenomenon. In this way base of the others, we can say, derived relations that are formulated concept that is based on the world – famous the result of the satisfaction of both sides’ needs. So, the gurus’ theories of the quality such as Deming,Juran, needs of the creator and of the consumer of the concrete Ishikava ,etc., is not exclusive , that means that it product define/generate the level of the same. The analyses critically the results of the well-developed consumer expresses his/her needs through demands and disciplines research concerning the quality , integrate the creator identifies them: (i) directly from the consumer them or reject them, proving in that way its inter- and/or (ii) through his/her own research of the needs disciplinary or pragmatic nature. Precisely speaking ,in trying to show his/her creative potentials through the this way formulated concept has in its essence the quality quality level that is upper than the consumer expects. as generally accepted paradigm of running the This is the point of the quality spiral that is moved by the organization. collective effort of the humanity for the development and The existing models of quality ,in the first place the prosperity with the aim of achieving the absolute limit, models known as TQM and QMS model that is based on that, speaking the quality language, me recognize as the the processes , are complex management concepts based quality of life. The processes of the research that move on the continuous progress of the quality of the entity that the quality spiral and that are its inalienable part, are is question and everything is in the domain of its different. The differences mainly refer to: (i) approach (ii) concretely set goals. These goals are the result of the methods and (iii) techniques and they are results of: (i) process of planning that will be reached by involving the available resources of the organization through adequate

125 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005 production process / services . The achieved results (oriented to logic and verbal expression )and right affirm themselves through the process of measurement / (capacity for spontaneity, intuition and meditation in assessment that results in information about the achieved pictures)cerebral hemisphere /3/ .Therefore, mental level of quality. Since the task of the management is not models should express the full complexity of real system only provided (fixed) quality , but also the but , because of the presence of the already mentioned improved/advanced one, then the measurement system, unclearness /irregularities , the way to the that is the evaluation must be used for establishing the goal is possible through the process of continuous achieved level of quality , and on the said spiral it adaptation to the situation. represents one step. In order to fix the existing level of quality of Since the processes of measurement , assessment ,and products that is of process and system ,the formation of evaluation are not “restricted” as terms ,then they are the model of assessment /evaluation of quality is directly connected with the problem of imprecision that unavoidable facing at the same time not only with the so makes the management process uncertain. called “hard “ data processing but also with the problem In order to improve the processes of measurement of “soft” data processing that refers first of all to the /assessment/evaluation by solving the imprecision like values ,the principles , the challenges excellent and so the unavoidable and completely natural thing’4 for on .The said problems and their solutions are “an modeling of such structures ,it is necessary to develop excuse” for the development of the model of quality more adequate method (promote the existing one) , and evaluation that includes “mechanism” of “soft” data that is the main intention of this research. processing ,with clearly defined structure –elements and interrelations. It is very important to mention in this context that: measurement, assessment and evaluation are close in meaning but they are not identical processes , so 2.DESCRIPTION OF THE PROBLEM it is necessary to define clearly what and how something is measured ,what and how something is assessed, and Analyzing the evolution of the quality (the existing what evaluation means. models) from the phase of inspection and control to the So, for the concrete case, the subject of measurement complete management of the quality it can be concluded must be exactly “determined”, then measuring device, that in all phases of development the obligatory contents unit of measurement, as well as methods and techniques of every model are –measurement and control. The of measurement. The example defined in this way refers quality inspection (measurement, assessment, evaluation) exclusively to the situation when all formal conditions becomes more and more complex problem that is directly for measurement are satisfied-so, it is about extensive connected with the evolution of quality .Combining the quality. In the case of intensive quality, the basic various technical of quality with the practical experience measurement can’t be done , the quality control is carried we get the models that contain “mechanism “ for the out by using the assessment procedure. Similarly, the constant improvement . evaluation that includes identification and defining of the Those are so-called models of business excellent that goals, for example the process contains at the same time are based on TQM approach . the assessment in relation to the set goals. That is why The basic principles of quality on which developed the evaluation is considered to be the broader term than models of the management of the quality are based are : the measurement and the assessment .So, the evaluation -the customer’s progress is system of measurement that is of assessment of the -constant improvement of process process accepting the broadly set goals. The result of -involvement of all employees . evaluation is the complete ”picture” of results of the Set aside the principle of constant progress of process, process shown through the level of achieved in relation since the insistence to projected quality or precisely, the relative value of on the respect of the same “produces” the procedure quality in relation to referent standard (model). of establishing the achieved level of the quality of In order to formulate the progresses it is necessary to process. define clearly the goals on the basis of the evaluation The weak side in TQM approach is the continuous results, and that means: progress of quality (i)endless race,(ii) no one can a)performances of the products and their constantly “improve” and (iii) the achieved “optimums” characteristics: is followed by satisfaction and fatigue. b)performances of the processes and their The dilemma is whether and which model TQM can characteristics: be realized in practice and how ?/2 / c)dimensions that are used to express the values of So, development (continuous progress) became characteristics: endless process of improvement . Uncertainty, dilemmas, d)limit that the characteristics must reach; as well as paradoxes can’t be solved satisfactorily only e)time interval during which the goal should be by existing procedures .This is as it is, because the same reached; are not rational demand but the questions of: (i) wishes Within this research into the characteristics of quality, ;(ii) emotions ;(iii) doubts, (iv) intuition and so on .What that is defined as the complete characteristic of the is relevant in this context is the analogy with the life entity(product, process, system) , the same are classified as follows:

4Verner Hajzenbergov - principle uncertainty 126 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

*characteristics with the quantity character –whose case is a fact that lows which are valid in classical set quality can be evaluated by exact measurement theory and / or lows of classical relations are not valid at (accepting the formal conditions of measurement) and fuzzy case in total /6/. This problem is solved in new *characteristics with the quality character-whose approach generalized on real valued logic and real valued quality can be evaluated by the use of various evaluation set theory. methods. So, the quality of the entity (product, process, system)is expressed through the characteristics of quantitative and qualitative character with the different “ 4. NEW REAL VALUED LOGIC level of construction” . In order to fix the total quality and make the method clear it is necessary to “precise “ New real valued logic and real valued sets theory the result of the use of the numbers. The use of the develop as non planning result in solving opening numbers in the process of evaluation is, itself, very problem in interpreting fuzzy measure in fuzzy integral complicated procedure. This problem receives full for multi attribute decision /7/.New approach has same attention in accepting three different uses of numbers: ambitions and aims as it is the case with fuzzy logic and *as symbols or signs for identification; fuzzy relation (which special case is fuzzy value of *as symbols that show the place of the level of membership function in sets theory case). New is that it quality in the series of levels and works on consistent generalization for change from *as symbols that show quantitative relations among fuzzy approach consistent generalization at logic case the characteristic. means that all tautologies are preserved an / or The existing models of quality: contradiction of classical logic in the generalize case too. a)without exception include module (evaluation, Every low from classical sets theory are preserved as in a assessment, self-assessment) and they take the quality to case of real valued sets theory( as the relation can be be the complex technical / economic category treated as sets on arranged n - tuple, it means hat all lows b)total quality of entity express on the basis of the from classical relations are preserved) .In general case fixed quantitative and qualitative characteristics;(its we have added complement properties , which are components , aspects, attributes ) important for analyzing problem of quality , but they are c)Ponder approach ( ponder coefficients ) is used as not in contradiction with classical case as it is in technique of aggregation that way more important conventional fuzzy logic (exactly all famous fuzzy logic). aspects are given higher ponder and apposite . Conventional fuzzy logic are base on principle of true From the literature ,it is clear, the approaches based functionality /8/.True functionality is a property that the on ponder coefficients cannot adequate treat problems value of true complex expression can determine directly which have not additive nature. For example if we on the base of its components . Direct consequence of this analyze simple cases which for the entity are relevant principle are known true tables . In case multi – valued only two attributes , but we are interested for one and logic we wanted to tell same (for example case three other attribute. value {0, 1 ,1} we’ll have true table with multi line –3n This simple and natural demand we are not capable to 2 model by ponder coefficients! That was the motive for line) . The problem is that we can not preserve all introduction non additive approach , as fuzzy measure tautologies and contradictions at the same time . and fuzzy integral /4/. In the same time , it shows us that At new approach , the corresponding name – the nature of problem is linked for quality and the interpolation logic structure and principle of structural classical mathematics based on black – white approach ( functionality is defined on the symbolic ( syntactic) level, something is true or not true) is not adequate. In general independent by value of variable. This principle is same case , it is not sufficient information if the feature we as in a classical logic case ( and classical sets theory), as analyzed exist or not but it is looking for gradation in a multi –valued logic case, which treated in new way . (intensity) feature . That was the motive to introduce It is shown that principle of true functionality represents fuzzy logic and theory of fuzzy sets/5/,in domain quality only consequence of principle of structural functionality of problems . and it is important only in two – valued classical, but in general case it makes known problems in fuzzy logic /9/. This result make it possible for direct generalize of classical theories based on classical logic and/or classical 3.FUZZY LOGIC AND FUZZY RELATIONS sets theory (which means-classical relations in general case). As it is said one of mathematics key tools in treating Illustration of this is shown by example from fuzzy the quality is relation. Classical relation for needs of preference structure /10/. treating problems which connect quality , is non adequate After twenty years of work on generalization known , for simple reason , it does not possess possibility to classical results of preference structures /11/ by fuzzy treat gradation of relation. Fuzzy relations have aim to technique , we get results whish have been asserting that treat case when gradation of relation exist . Relation in was not possible to get them directly from classical unary relation make it possible to treat properties which results . In the paper we mentioned /10/, it is shown that have gradation , just that what is important for the new approach gives results in simple and natural way mathematics treating problem connected with quality. , directly. Just , this result is of great importance for Defect of fuzzy sets theory and fuzzy relations in general treating problems from domain quality . 127 Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005

5. NEW APPROACH EVALUATION IN /4/Grabisch, M. Alternative Representations of Discrete QUALITY Fuzzy Measures for Decision Making. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems 5(5) (1997) 587-608 New approach to the quality evaluation is based on /5/Zadeh, L. A. Fuzzy sets. Information and Control 8, 3 new real-valued logic. (1965), 338-353. Attribute of quality is treated as element of entity /6/Klement, P., Mesiar, R., Pap, L. Triangular norms, (analyzed system) which is analyzing from the aspect of volume 8 of Trends in Logic. quality we correspond attribute the norm value from real Kluwer, Dordrecht, (2000). interval [0,1]. What contributions of analyzing attribute /7/Radojevic, D. New [0, 1]-valued logic: A natural generalization of Boolean logic, Yugoslav Journal of quality is bigger its value will be near 1 and opposite will Operational Research – YUJOR, Belgrade, 10 (2, (2000), 185- be near 0. 216. Quality of entity determines by aggregation as logical /8/Hajek, P. Metamathematic of Fuzzy Logic, Kluwer or pseudo logical function of analyzing attributes Academic Publishers, (1998). (aspects) of quality. /9/Radojevic, D. Logical measure - structure of logical In case of logical function aspects which are analyzed formula, Technologies for Constructing Intelligent Systems 2, , by logical operator determined by nature of problem Physica- Verlag, (2002),417 -429. /10/Radojevic, D. Fuzzy preference structures based on more precisely it is determined by relations from analyst 3 to the analyzed system . On that way, the analyzed Syntactic Structured and Semantic Convex (S C) logic, EUROFUSE 2002, Varenna, Italy (2002), 231-235 aspects whose nature in practice is heterogeneous , are /11/Arrow, K. J. Social Choice and Individual Values, treated consistently trough quality, therefore , uniform Wilez, New York, (1951). visum (independently from nature of analyzed aspect). In the pseudo-logical aggregation case we have combination of logical terms by attributes and corresponding function with ponder coefficients. Well , usually ,approach in aggregation by ponder coefficients which are corresponding to analyzed aspects ( ponder- approach) which is in dominates use in today practice is only one special case of this other approach this disconnect any logical connection of attributes . That is the mean reason why existing approach are mostly inadequate in real problems.

CONCLUSION

In this paper came into view for apply generalized logic at problems which are connected at the quality evaluation in composite system based in the quality of relevant aspects ( attributes , characteristics , index). Classical logic can not treat nuance –gradation . Fuzzy logic has possibility to treat gradation. Elementary defect of fuzzy logic is in the same frame in which the classical logic is, fuzzy logic cannot preserve all tautologies from classical logic.That can be a problem . In this paper , new real valued logic is proposed as a tool for treating problem at the quality evaluation . New real valued logic is consistent generalization of classical logic and it possess new complementary features whish are important at the quality evaluation. Next papers will be illustrated by the real examples .

REFERENCES

/1/MajstorovićV., One TQM Model (on Serbian), Bussines Politic,Belgrade (2000) /2/Stanivuković D., From QMS to TQM, Proceedings of Quality Conference, Kragujevac (2003), 273-279 /3/Schwaninger M., A Practical Approach to Strartegy Development ,Long Rang Planning, No 5.,(1987)

128

University of Belgrade – Mechanical Engineering Faculty Laboratory for Production Metrology and TQM

FINAL REPORT FOR CIRP:

Third International Working Conference

“TOTAL QUALITY MANAGEMENT – ADVANCED AND INTELLIGENT APPROACHES”

WAS HELD 30th May – 1st June, 2005, Belgrade, Serbia & Montenegro

Report prepared by:

Professor Dr. Vidosav D. MAJSTOROVIC Conference Chairman

CIRP member support Prof. Dr. Albert Weckenmann

Belgrade, July 2005.

129 Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

C O N T E N T S

I. General data…………………………………………………………………………………………………………………. iii

II. Paper sessions ……………………………………………………………………………………………………………. iii

Appendix I: Lists of titles and authors and sessions chairmen …………………………………………. v

III. Financial data ……………………………………………………………………………………………………………. xi

IV. Concluding notes ……………………………………………………………………………………………………… xi

V. ROUND TABLES WAS HELD ON 3rd IWC TQM 2005 ………………………………………………….. xii

Appendix II. Special edition International Journal “ TOTAL QUALITY MANAGEMENT & EXCELLENCE“ ………………………………………………………………………………………………………………………. xiv

Appendix III. Fourth International Working Conference - FIRST ANNOUNCEMENT - CALL FOR PAPERS, PRESENTATIONS AND PARTICIPATION...... xxiv

130 ii Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

I. General data:

1. Name of the Conference: Third International Working Conference: “ Total Quality Management – Advanced and Intelligent Approaches “ 2. Date and place: 30th May – 1st June, 2005, Belgrade, Serbia & Montenegro 3. Number of participants: 241 from 18 countries ( Australia, Austria, Germany, France, England, Slovenia, Italia, Macedonia, Greece, Japan, Indonesia, The Netherlands, Sweden, Canada, Romania, Republika Srpska, Ukraine, Serbia & Montenegro )

II. Paper sessions:

1. Number of papers received: 53 2. Number of papers accepted: 37 3. Number of papers presented: 34 4. Number of sessions and names: 1 Opening Plenary Session, 6 Paper Sessions and 2 Round Table Discussion.

CONFERENCE PLAN

30th May, 2005 / Monday 16:00 – 17:30 Registration on reception desk (first floor) in Palace of Syndicate, (Main entrance - Square Nikola Pašić), str. Dečanska 14, Belgrade. 17:00 Welcome Cocktail in Palace of Syndicate, (Main entrance - Square Nikola Pašić), str. Dečanska 14, Belgrade / Club Promotion (ground level). 19.00 Round Table Disscussion: ASSESSING QM LEVEL – FUTURE DEVELOPMENT TRENDS – Siemens, Belgrade.

31st May, 2005 / Tuesday 08:00 – 17:30 Registration on reception desk in Palace of Syndicate, Hall 2 (first floor), Square Nikola Pašić, Belgrade. 09:00 – 10.30 Opening Ceremony and Plenary Presentation - Palace of Syndicate, Hall 2 (first floor) - Square Nikola Pašić, Belgrade. 10:30 – 11:30 Cocktail - Palace of Syndicate, Club Promotion (ground level), Square Nikola Pašić, Belgrade. 11:30 – 13:00 Opening Plenary Session: IMS, TQM and BE / State of the Art - Palace of Syndicate, Hall 2 (first floor) - Square Nikola Pašić, Belgrade. 13:00 – 14:30 Break 14:30 – 16:00 Parallel Sessions (3) - Palace of Syndicate, (Main entrance - Square Nikola Pašić), str. Dečanska 14, Belgrade / Hall 2 (first floor), Hall 3 (first floor), Hall 4 (ground level). 16:00 – 16:30 Coffee break - Club Promotion (ground level).

131 iii Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

16:30 – 18:00 Parallel Sessions (3) - Palace of Syndicate, (Main entrance - Square Nikola Pašić), str. Dečanska 14, Belgrade / Hall 2 (first floor), Hall 3 (first floor), Hall 4 (ground level). 20:00 Conference Dinner (informal).

1st June, 2005 / Wednesday . 20.00 Round Table Disscussion: Integrated Management Systems - State-of-the- art and future trends, Club of Railway Health Care Institute, Str. Savska 23 Belgrade

5. Time reserved for each paper: 30 minutes for Keynote papers and 20 minutes for other papers 6. Time reserved for discussion: 10 minutes for Keynote paper and 5 minutes for other papers 7. Attendance to sessions: Average 200 participants per session

8. Members of Scientific Committee

International Program Committee: Honorary Chairs: prof. Dr. G. Arndt (AU), M. Debenham (UK), prof. Dr. S. Karapetrovic (CA), prof. Dr. L. Monostori (H), prof. Dr. H. Osanna (A), prof. Dr. T. Pfeifer (G), prof. Dr. G. Sohlenius (S), prof. Dr. H. VanBrussel (B), prof. Dr. A. Weckenmann(G), prof. Dr. A. van der Wiele (NE).

Conference Chair: Prof. Dr. V. Majstorovic (S&M).

Members: prof. Dr. H. Bley (G), prof. Dr. P. Bertok (AU), prof. Dr. P. Bojanić (S&M), prof. Dr. C. Bouzakis (GR), prof. D. Brissaud (F), prof. Dr. Chryssolours (Gr), prof. Dr. K. Cho (K), prof. Dr. L. DeChiffre (DK), B. Dimitrijević (CA), Dr. C. Evans (USA), prof. Dr. G. Halevi (IS), prof. Dr. T. Hardjono (NL), prof. Dr. S. Ho (HK), B. Jouslin (B), prof. Dr. J. Jedrzejewski (PL), prof. Dr. Z. Katz (SA), prof. Dr. A. Kjellberg (S), prof. Dr. V. Milačić (S&M), M. Mlađenovic (Ca), prof. Dr. P. Molnar (H), prof. Dr. G. Morel (F), prof. Dr. S. Nof (USA), S. Olivas (SP), prof. Dr. V. Parežanin (S&M), Dr. G. Pegs (UK), Mr. Z. Pendić (S&M), prof. Dr. S. Sartori (I), Mr. F. Steer (UK), prof. Dr. M. Soković (Sl), prof. Dr. D. Stanivuković (S&M), prof. Dr. R. Teti (I), prof. Dr. J. Rogerson (UK), Mr. S. Živković (S&M), prof. Dr. S. Yamada (J). Note: Bolded name are CIRP member

Sponsored by: International Institution for Production Engineering Research (CIRP), Paris, France.

Co– sponsored by: Japanese Union of Scientits and Engineers (JUSE), Tokyo, Japan. European Organization for Quality (EOQ), Brusssels, Belgium. European Foundation for Quality Management (EFQM), Brussels, Belgium. American Society for Quality (ASQ), Milwoku, USA.

132 iv Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

International Federation for Information Processing (IFIP), Laxenburg, Austria. International Measurement Confederation (IMEKO), Budapest, Hungary. International Federation of Automatic Control (IFAC), Laxenburg, Austria.

Organized by Mechanical Engineering Faculty, Laboratory for Production Metrology and TQM, Belgrade, Serbia & Montenegro.

9. Members of the Organizing Committee: Prof. Dr. V. Majstorovic, T. Stojiljkovic, G. Dojcinovic 10. Short description of the paper reviewing procedure: This procedure was based on CIRP evolution form

Appendix I: Lists of titles and authors and sessions chairmen:

DETAILED PROGRAM

30th May, 2005 / Monday 16:00 – 17:30 Registration on reception desk (first floor) in Palace of Syndicate, (Main entrance - Square Nikola Pašić), str. Dečanska 14, Belgrade. 17:00 Welcome Cocktail in Palace of Syndicate, (Main entrance - Square Nikola Pašić), str. Dečanska 14, Belgrade / Club Promotion (ground level). 19.00 Round Table Disscussion: ASSESSING QM LEVEL – FUTURE DEVELOPMENT TRENDS – Place will be definition later (information will be delivery on request), Belgrade.

31st May, 2005 / Tuesday 08:00 – 17:30 Registration on reception desk in Palace of Syndicate, Hall 2 (first floor), Square Nikola Pašić, Belgrade. 09:00 – 10.30 Opening Ceremony and Plenary Presentation - Palace of Syndicate, Hall 2 (first floor) - Square Nikola Pašić, Belgrade. • Opening addresses • Awards • Opening Plenary Presentation – EOQ – VIEW IN FUTURE, Mr. Bertrand JOUSLIN, General Secretary of EOQ, Brussels, Belgium 10:30 – 11:30 Cocktail - Palace of Syndicate, Club Promotion (ground level), Square Nikola Pašić, Belgrade.

133 v Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

11:30 – 13:00 Opening Plenary Session: IMS, TQM and BE / State of the Art - Palace of Syndicate, Hall 2 (first floor) - Square Nikola Pašić, Belgrade.

Chairmen: Prof. Dr. Herbert Osanna, Vienna University of Technology, Vienna, Austria. Prof. Dr.-Ing. Albert Weckenmann, Chair Quality Management and Manufacturing Metrology, Universuty Erlangen-Nuremberg, Erlangen, Germany. Prof. Dr. Ton van der Wiele, Erasmus University Rotterdam, Rotterdam, The Netherlands. Michael Debenham, Professional Affairs Manager, Institute of Quality Assurance, London, UK. Prof. Dr. Stanislav Karapetrović, University of Alberta, Edmonton, Canada. Prof. Dr. Vidosav D. Majstorović, Faculty of Mechanical Engineering, Belgrade University, Belgrade, Serbia. Prof. Dr. Dragutin Stanivuković, FTS, University of Novi Sad, Novi Sad, Serbia.

Keynote Speakers 1. INTELLIGENT FLEXIBLE DISASSEMBLY AND RECYCLING OF USED PRODUCTS TO SUPPORT TOTAL QUALITY MANAGEMENT AND SUSTAINABILITY IN EUROPEAN INDUSTRY, Keynote paper, Prof. Dr. Herbert Osanna, Dr. M. N. Durakbasa, H. S. Tahirova, Vienna University of Technology, Vienna, Austria. 2. COMPUTER - AIDED DESIGN OF EXPERIMENTS FRAMEWORK - A COMPREHENSIVE APPROACH TO PROCESS IMPROVEMENT, Keynote paper, Prof. Dr. Albert Weckenmann, Peter-Frederik Brenner, Chair Quality Management and Manufacturing Metrology, Erlangen University Erlangen-Nuremberg. 3. A MANAGEMENT CONTROL PERSPECTIVE FOR QUALITY MANAGEMENT: AN EXAMPLE IN THE AUTOMOTIVE SECTOR, Keynote paper, Jos van Iwaarden, Prof. Dr. Ton van der Wiele, Erasmus University Rotterdam, The Netherlands. 4. THE VALUE OF ORGANISATIONAL CULTURE AND THE ROLE OF COMPETENCIES IN DELIVERING QUALITY PRODUCTS AND SERVICES, Keynote paper, Michael Debenham, Professional Affairs Manager, Institute of Quality Assurance, London, UK. 5. IMS IN THE M(E)SS WITH CSCS, Keynote paper, Prof. Dr. Stanislav Karapetrović, Auditing and Integration of Management Systems Research Laboratory, Department of Mechanical Engineering, University of Alberta, Edmonton, Canada. 6. A CONTRIBUTION TO THE DIGITAL QUALITY CONCEPT RESEARCH, Keynote paper, Prof. Dr. Vidosav D. Majstorović, Faculty of Mechanical Engineering, Belgrade University, Belgrade, Serbia.

14:30 – 16:00 Session 1 Palace of Syndicate / Hall 2 (first floor). Session 1 - Organizational and Bussines Excellence.

Chairmen: Dr. Raine Isaksson, Gotland University, Visby, Sweden. Yagodzinsky Viktor, Ph.D., Personnel Training Centre “PRIROST-Academy”, Kiev, Ukraine Mr Milan Ivanović, LR, Belgrade Dr. Ivan Skubic, Metrology Institute of the Republic of Slovenia (MIRS), Ljubljana, Slovenia. Prof. Dr. Miroslav Bobrek, Faculty of Mechanical Engineering, University of Banjaluka, Republika Srpska. Prof. Dr. Zivadin Micić, Techical Faculty in Čačak, University of Kragujevac, Srbia.

134 vi Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

1. ORGANISATIONAL SUSTAINABILITY MANAGEMENT TRROUGH MINIMISED BUSINESS EXCELLENCE MODELS, Keynote paper, Rickard Garvare, Luleĺ University of Technology, Luleĺ, Sweden , Dr. Raine Isaksson, Gotland University, Visby, Sweden. 2. PERSONNEL TRAINING AS ASPECT OF INTEGRAL BUSINESS EXCELLENCE, Yagodzinsky Viktor, Ph.D., Personnel Training Centre “PRIROST- Academy”, Kiev, Ukraine. 3. QUALITY MANAGEMENT MATURITY AND MANAGEMENT ATTITUDE TO QUALITY, 1) MSc Milan Ivanović, 2) Prof. Dr. Vidosav D. Majstorović, 1)LR, Belgrade 2)Mechanical Engineering Faculty, Belgrade. 4. EXCELLENCE IN PUBLIC ADMINISTRATION - EXPERIENCES AND EXAMPLE OF MIRS, Dr. Ivan Skubic, Mag. Karmen Kern Pipan, Metrology Institute of the Republic of Slovenia (MIRS), Ljubljana, Slovenia. 5. QUALITY MANAGEMENT SYSTEM IN LOCAL ADMINISTRATIONS, Ljiljana V. Radovanović, City administration of Banjaluka, Head of IT sector, Prof. Dr. Miroslav Bobrek, Faculty of Mechanical Engineering, University of Banjaluka, Republika Srpska. 6. TQM MODELLING AND IMPROVING QUALITY OF INFORMATION TECHNOLOGY ON RASINA DISTRICT SCHOOLS AS EXAMPLES, Milica Tufegdzic, Machine-electrotehnical school in Krusevac, Prof. Dr. Živadin Micic, Techical Faculty in Čačak, University of Kragujevac, Srbia.

15:00 – 16:30 Session 2 – Palace of Syndicate / Hall 3 (first floor). Session 2 – Intelligent Metrology and Intelligent Manufacturing.

Chairmen: Prof. Dr. Mirko Sokovic, Faculty of Mechanical Engineering, University of Ljubljana,Slovenia. Jens Dören, Fraunhofer Institute for Production Technology, Department Metrology and Quality Management, Aachen, Germany. Dr. G. Andreadis, Laboratory for Machine Tools and Manufacturing Engineering, Mechanical ; Eng. Dept., Aristoteles University of Thessaloniki, Greece. Doc. Dr. Miodrag Stevic, Insitute for Production Engineering, Faculty of Engineering, University of Novi Sad, Novi Sad, Serbia and Montenegro. Prof. Dr. Bojan Ačko, Mechanical Engineering Faculty, University of Maribor, Slovenia. MSc. Srđan Živković, Aeronautical Institute, Belgrade, Serbia.

1. DESIGN OF A SYSTEM FOR INTELLIGENT DATA-POINT PRE-PROCESSING IN REVERSE ENGINEERING, Keynote paper, MSc. I. Budak a), M. Sokovic b), J. Hodolic a),a) Faculty of Technical Sciences, University of Novi Sad, Serbia, b) Faculty of Mechanical Engineering, University of Ljubljana,Slovenia. 2. „FAST FORWARD” – DESIGNED PROCESS OPTIMISATION, Prof. Dr.-Ing. R. Schmitt; Dipl.-Ing. J. Dören, Fraunhofer Institute for Production Technology, Department Metrology and Quality Management, Aachen, Germany. 3. INTERNET BASED AUTOMATION OF THE PROCESS "DESIGNING - MACHINING" OF A WORKPIECE, Prof. Dr. K.-D. Bouzakis(1), Assistant Prof. Dr. A. Vakali(2), Lecturer Dr. G. Andreadis(1), E. Karapidakis(2), 1 Laboratory for Machine Tools and Manufacturing Engineering, Mechanical ; Eng. Dept., Aristoteles University of Thessaloniki, Greece; 2 Computer Science Dept., Aristoteles University of Thessaloniki, Greece. 4. DEVELOPMENT OF AN INTELLIGENT MODULE FOR DECREASE OF MEASURING ERROR ON CMM, M. Stevic1, J. Hodolic1 S. Vukmirovic2 , 1Insitute for Production Engineering, Faculty of Engineering, University of Novi Sad, Novi Sad, Serbia and Montenegro;

135 vii Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

2Department of Control Systems and Automation, Faculty of Engineering, University of Novi Sad, Novi Sad, Serbia. 5. TRACEABILITY OF MEASUREMENT RESULTS IN INDUSTRY, Prof. Dr. Bojan Ačko, Mechanical Engineering Faculty, University of Maribor, Slovenia 6. POSITION ERROR’S INSPECTION APPLIED ON FREE FORM SURFACES, MSc. Srđan Živković, Zoran Tumbas, Aeronautical Institute, Belgrade, Serbia.

15:00 – 16:30 Session 3 – Palace of Syndicate / Hall 4 (ground level). Session 3 – World Class Performance.

Chairmen: Dr. Lukman Sukarma, Agency for the Assesment and Application of Technology, Indonesia. MSc. Zoran Pendić, Institute of Nuclear Sciences “VINCA”, Belgrade, S&M. (Mrs) Slavka Backović – Jeremić, M.E., Zastava arms, Kragujevac, Serbia. Branko Dimitrijevic, Mech. Eng., North Shore Management Systems Inc., AQSR Windsor, Canada.

1. SYNERGY IN APPLICATION OF PRACTICES OF THE INTEGRATED PRODUCTION SYSTEM TOWARDS ACHIEVING MANUFACTURING EXCELLENCE, Keynote paper, Dr. Lukman Sukarma, Agency for the Assesment and Application of Technology, Indonesia. 2. CONTINUOUS IMPROVEMENT – A PREREQUISITE FOR SUCCESS, Keynote paper R. Pendić1, V. Majstorović2, Z. Pendić3,1 Eurosystems Group – Matheos Invest Group, Belgrade, S&M; 2 Mechanical Engineering Faculty, Belgrade University, S&M; 3 Institute of Nuclear Sciences “VINCA”, Belgrade, S&M. 3. ACCELERATING PERFORMANCE IMPROVEMENT THROUGH APPROPRIATE RESOURCE ALLOCATION, Dr. Lukman Sukarma, Agancy for the Assessment and Aplication of Technology, Indonesia. 4. TQM – A CONDITION OF COMPETITEVENESS AT OPEN MARKET, Slavka Backović – Jeremić, M.E., Zastava arms, Kragujevac, Serbia. 5. TOYOTA PRODUCTION SYSTEM – PRINCIPLES AND METHODOLOGIES, B. Dimitrijevic, Mech. Eng., North Shore Management Systems Inc., AQSR Windsor, Canada.

16:00 – 16:30 Coffee break - Club Promotion (ground level).

16:30 – 18:00 Session 4 – Palace of Syndicate / Hall 2 (first floor). Session 4 - Bussines Process Improvement.

Chairmen: Prof. Dr. Vojislav Stoiljković, CIM College d.o.o, Niš, Srbia. Prof. Dr. Y. Sujar , Mechanical Engineering Faculty, Tamilnadu, India. Dr. Tolnay Marián, Slovak University of Technology in Bratislava Faculty of Mechanical Engineering, Bratislava, Slovak Republic. MSc. Zoran Pendić, Institute of nuclear sciences "Vinča", Belgrade, Serbia. (Ms) Tatjana V. Šibalija, TMicroelectronics Malta Ltd., Industry Road, Kirkop, Malta.

1. SIX SIGMA LOAN-PROCESSING PROCESS, Keynote paper, Prof. Dr. Vojislav Stoiljković, Predrag Stoiljković, CIM College d.o.o, Niš, Srbia. 2. METHODOLOGY FOR ACHIEVING PROCESS IMPROVEMENT GOALS IN SOFTWARE INDUSTRIES, *Dr.P.Balachandran **Mr.Y.Sujar **Mr.N.Ramasamy, * Scientist, LPSC,

136 viii Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

Thiruvanantha puram, Kerala.India **Assistant professor, Mechanical Engg., Noorul Islam College Of Engg., Kumara Koil, Tamilnadu, India. 3. INTEGRATED SLOVAK SCIENCE IN EUROPEAN RESEARCH SPACE, Dr.Tolnay Marián, Slovak University of Technology in Bratislava Faculty of Mechanical Engineering, Bratislava, Slovak republic 4. WORK SAFETY POLICY IN HEALTH INSTITUTIONS – WITH THE PARTICULAR REFERENCE TO WORK SAFETY OF THE MRI CENTER, Zoran Pendić1, Biljana Jakovljević2, Jasmina Stevanović2, Rajko Pendić3, Bojana Jakovljević3, 1Institute of nuclear sciences "Vinča", Belgrade; 2Institute of oncology and radiology of Serbia, Belgrade; 3Eurosystems Group - Matheos Invest Group, Belgrade, Serbia. 5. APPLICATION OF TAGUCHI MODEL FOR QUALITY PRODUCT IMPROVEMENT, Mech. Eng. B.Sc. Tatjana V. Sibalija1, Prof. Dr. Vidosav D. Majstorovich2 ,1 TMicroelectronics Malta Ltd., Industry Road, Kirkop, Malta, 2 Mechanical Engineering Faculty, Laboratory for Production Metrology and TQM, Serbia.

16:30 – 18:00 Session 5 – Palace of Syndicate / Hall 3 (first floor). Session 5 – Intelligent Quality Tools.

Chairmen: Laura Catellani, Dipartimento di Meccanica – Politecnico di Milano, Milano, Italy. Mr. Zoran Obradović, EPS Center for Quality, Beograd, Serbia. S. Randjelovic, CIM TTC laboratory, Faculty of Mechanical Engineering Nis, Serbia. P.Stanaćev, DAIDO METAL KOTOR AD, Kotor, Serbia & Montenegro. Siniša Marković, Mechanical Engineering Faculty, Belgrade, Serbia. Srđan Gligorić, Mechanical Engineering Faculty, Belgrade, Serbia.

1. USING PARTIAL LEAST SQUARES REGRESSION TO ANALYSE QUALITY IN HIGHER EDUCATION, Keynote paper, Laura Catellani, Bianca M. Colosimo, Q. Semeraro, Dipartimento di Meccanica – Politecnico di Milano, Milano, Italy. 2. APPLICATION OF PARETO METHOD IN SERBIAN ELECTRIC-POWER COMPANY (EPS), Prof.dr Vojislav Stoiljković 1), Mr. Zoran Obradović 2), Bratislav Stoiljković 3), 1) CIM College d.o.o., Niš; 2)EPS Center for Quality, Beograd; 3) CIM College d.o.o. Niš, Serbia. 3. PRODUCTION OF ALUMINIUM STRUCTURE WITH EXTRUSION TECHNOLOGY SUPORT QFD METHOD, S. Randjelovic*, P. Milosavljevic, S. Mladenovic, *CIM TTC laboratory, Faculty of Mechanical Engineering Nis, Serbia. 4. QUALITY AND ENVIRONMENT IMPROVEMENT - ONE INDUSTRIAL EXPERIENCE, P.Stanaćev1), A.Tasovac1), N.Kaluđerović1), G. Radović1) , Y.Kawamura2), .Kitagawa2), K.Shigemitsu2), 1) DAIDO METAL KOTOR AD, Kotor, Serbia & Montenegro, 2) DAIDO METAL COMPANY Ltd, Nagoya, Japan. 5. SOFTWARE FOR BUSINESS EXCELLENCE ASSESSMENT OF SMALL AND MEDIUM ENTERPRISES, Siniša Marković, Mechanical Engineering Faculty, Belgrade, Serbia. 6. KNOWLEDEGE BASED SYSTEM FOR CMM INSPECTION PLANING, Srđan Gligorić, Mechanical Engineering Faculty, Belgarde, Serbia.

16:30 – 18:00 Session 6 – Palace of Syndicate / Hall 4 (ground level). Session 6 – Integrated Managament Systems.

137 ix Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

Chairmen: MSc Nenad Stefanovic, Zastava Automobiles, Kragujevac, Serbia. Prof. Dr. Pavel Castka, University of Canterbury, Christchurch, New Zealand. MSc. Ilija Đekić, AQA International - Eastern Europe, Belgrade, Serbia. MSc. Marko Mirković, Bauxite mines,s.c. Quality & Information Centre, Niksic, Montenegro. PhD. Dragan Ranđelović, Technical Faculty of Bor, Bor, Serbia.

1. RESEARCH AND DEVELOPMENT OF DIGITAL QUALITY MODEL IN SCM, Keynote paper, 1) Nenad Stefanovic, Zastava Automobiles, 2) Vidosav Majstorovic, 3) Dusan Stefanovic, 1)Information Systems Division, Kragujevac,2)Mechanical Engineering Faculty, Belgrade, Serbia; 3)Faculty of Science, Kragujevac, Serbia. 2. ISO MANAGEMENT SYSTEMS STANDARDS AND SOCIAL RESPONSIBILITY CONNECTION: (NOT QUITE) JOINED-UP OPINIONS OF ISO’S STAKEHOLDERS, Pavel Castka 1), Michaela A. Balzarova 2) 1) University of Canterbury, Christchurch, New Zealand, 2) Brno University of Technology, Brno, the Czech Republic. 3. INTENDED AND UNINTENDED OUTPUTS IN INTEGRATED MANAGEMENT SYSTEMS, Ilija Đekić, M.Sc.Mech.E., Maja Simov, B.Sc., AQA International - Eastern Europe, Belgrade, Serbia. 4. RESEARCH INTO THE POSSIBILITIES OF THE APPLICATION OF THE FUZZY LOGIC TO THE DEVELOPMENT OF THE QUALITY EVALUATION MODEL, Marko Mirkovich, M.Sc.1), Prof. Dr. Dragan Radojevich2) , Prof. Dr. Vidosav D. Majstorovich3), Prof. Dr. Janko Hodolic4), 1 Bauxite mines,s.c. Quality & Information Centre,2 Mihajlo Pupin Institute, Belgrade, 3 Mechanical Engineering Faculty, Laboratory for Production Metrology and TQM, 4 FTS, Production Mechanical Engineering Institute, Novi Sad, Serbia. 5. INFORMATIONAL SYSTEM MODEL FOR MANAGING OF BUSINESS ACTIVITIES, PhD. Dragan Ranđelović, Zvonko Damnjanović, Vojkan Nikolić, Technical Faculty of Bor, Bor, Serbia.

20:00 Conference Dinner (informal).

1st June, 2005 / Wednesday

20.00 Round Table Disscussion: Integrated Management Systems - State-of-the-art and future trends, Club of Railway Health Care Institute, Str. Savska 23

138 x Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

III. Financial data

Registration fee: For IWC TQM registration fee was EURO 100. Participants from S&M was paid EURO 40. CIRP members (authors and IPC members not paid fee ). Sponsors representatives from S&M not paid fee. Actually we have small profit form this Conference. But I was paid CIRP fee for this Conference 300 Euro. Financial data details INCOME: Conference fee 106 participants – 5900 EUR, Sponsorship - 2300 EUR, Other income 1950 EUR. TOTAL “A” 10150 EUR EXPENSES: Cost for Conference preparation – 2050 EUR, Printing Materials – 3850 EUR, Running Costs – 2200 EUR, Other – 850 EUR. TOTAL “B” 9950 EUR HENCE PROFIT “P”= A-B= 200 EUR Note: CIRP FEE 300 Euro

IV. Concluding notes

General considerations on the quality of the papers and presentation Very good papers with advanced approaches from main Conference topics.

General consideration on the success of the Conference Conference is full success and Four IWC TQM & AIA will be held 2007. (Please see appendix III). Difficulty More applicable information for participants from industry.

139 xi Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

V. ROUND TABLES WAS HELD ON 3rd IWC TQM 2005

Third International Working Conference TOTAL QUALITY MANAGEMENT – ADVANCED AND INTELLIGENT APPROACHES, Belgrade, Serbia. E-mail: [email protected]

ROUND TABLE DISCUSSION 1

Meeting with cocktail to be held on 2005 – 05 – 30 7 p.m. – 9 p.m. Siemens d.o.o. Beograd, str. Omladinskih Brigada 21/III, Belgrade.

You are kindly invited to participate on the RTD 1 to be held in Siemens d.o.o. Beograd, str. Omladinskih Brigada 21/III.

AGENDA

1. Opening. Welcome (Prof. Dr. V. Majstorović) 2. Introduction of present and guests 3. Discussion – The assessment of QM level - State-of-the-art and future trends (Panelists: Prof. Dr. Ton Wiele, Prof. Dr. Herbert Osanna, Mr. Michael Debenham, Mr. Milan Ivanović, Mrs. Laura Catellani, Mr. Miodrag Vuković). In discussion was participate: all panelist, V. Majstorović, V. Simić, M. Minić, M. Jelić, P. Chatska, D. Stanivuković. 4. Cocktail

On meeting are invited and attending:

A. Professors and QM experts form abroad, participants on 3rd IWC TQM, BCQ, ICQ - Belgrade 2005 . B. National experts for QM.

In Belgrade, 30th May 2005. Prof. Dr. Vidosav Majstorović

Attending 58 participants for 10 countries. RTD was finished in 9.30 p.m.

Third International Working Conference TOTAL QUALITY MANAGEMENT – ADVANCED AND INTELLIGENT APPROACHES, Belgrade, Serbia. E-mail: [email protected]

140 xii Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

ROUND TABLE DISCUSSION 2

Meeting with dinner to be held on 2005 – 05 – 31 6 p.m. – 8 p.m. Club of Railway Health Care Institute, Str. Savska 23 Belgrade

You are kindly invited to participate on the RTD 2 to be held in Belgrade, Club of Railway Health Care Institute, Str. Savska 23.

AGENDA

1. Opening. Welcome (Prof. Dr. V. Majstorović) 2. Introduction of present and guests 3. Discussion – Integrated Management Systems – State-of-the-art and future trends (Panelists: Prof. Dr. Stanislav Karapetrović, Mr. Michael Debenham, Dr. Raine Isaksson, Mrs. Liliana Nitu, Prof. Dr. Miroslav Bobrek) In discussion was participate: all panelist, V. Majstorović, L. Nity, M. Soković, P. Chatska, Z. Pendić, M. Bulatović.

4. Dinner

On meeting are invited and attending:

A. Professors and QM experts form abroad, participants on 3rd IWC TQM, BCQ, ICQ - Belgrade 2005. B. National experts for QM.

In Belgrade, 31st May 2005. Prof. Dr. Vidosav Majstorović

Attending 46 participants for 12 countries. RTD was finished in 11.30 p.m.

141 xiii Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

Appendix II. Special edition International Journal “TOTAL QUALITY MANAGEMENT & EXCELLENCE “, Vol. 33, No. 3, 2005, Belgrade, Serbia.

ASSOCIATION SERBIA & MONTENEGRO FOR QUALITY AND STANDARDS Kneza Miloša 9, 11000 Beograd, Serbia, Phone/fax: ++ 381 11 323 55 15

INTERNATIONAL JOURNAL

TOTAL QUALITY MANAGEMENT & EXCELLENCE

'' Special edition ''

Papers second reviewed and presented at the Third International Working Conference ''Total Quality Management – Advanced and Intelligent Approaches'', held from 30th May to 1st June, 2005, at Belgrade, Serbia.

UDK 658.5 YU ISSN 1452-0680 No 3. Vol. 33 2005

142 xiv Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

INTERNATIONAL JOURNAL „TOTAL QUALITY MANAGEMENT & EXCELLENCE” Vol. XXXIII, No.3, 2005

FOUNDER AND PUBLISHER: Association Serbia and Montenegro for Quality and Standards (YUSQ), Belgrade

EDITOR IN CHIEF: Prof. Dr Vidosav D. Majstorović, mech. eng. Mechanical Engineering Faculty, University of Belgrade, Serbia

INTERNATIONAL EDITORIAL BOARD: Prof. dr Guenter ARNDT Prof. dr Gunnar SOHLENIUS University of Wollongong, Wollongong, Royal Institute of Technology, Stockholm, Australia Sweden Prof. dr Daniel BRISAUND Prof. dr Dragutin STANIVUKOVIĆ University of Grenoble, Grenoble, France FTN, Novi Sad, S&M Prof. dr Barrie DALE Prof. dr Herbert OSANNA UMIST, Manchester, England Technical University, Wien, Austria Michel DEBENHAM Prof. dr Tilo PFIEFER Institute of QA, London, England RWTH Aachen, Aachen, Germany Sofija DJURDJEVIC Prof. dr Hendrik VanBRUSSEL YUQS, Belgrade, S&M Katolike University Leuven, Heverlez, Belgium Prof. dr Noriaki KANO Prof. dr Shu YAMADA Science University of Tokyo, Tokyo, Japan Science University of Tokyo, Tokyo, Japan Prof. Dr. Laslo MONOSTORI Prof. dr Albert WECKENMANN Hungarian Academy, Budapest, Hungary University Erlangen, Erlangen, Germany Prof. dr Vidomir PAREŽANIN Prof. dr Ton ven der WIELE ASTRA grupa, Beograd Erasmus University Roterdam, The Mr Zoran PENDIĆ Netherlands Institut Vinča, Beograd

TECHNICAL EDITOR: Siniša M. MARKOVIĆ, Dipl. mech. eng.

PROOFREADING Printing from authors' disketes

EDITOR’S ADDRESS: Association Serbia & Montenegro for Quality and Standards, Serbia and Montenegro, 11000 Beograd, Kneza Miloša st. 9, Phone/fax: ++ 381 63 33 10 15 E-mail: [email protected] Web page: www.jusk.org.yu

This edition financially aided by the Ministry for Science, Technology and Development of the Republic of Serbia. Manuscripts and illustrations not returned. The Journal is exempted from taxes, Decree No.413.304/7502 of 4. June 1985, passed by the Republican Secretariat for Culture. The texts published in this edition of the JOURNAL cannot be copied or printed without prior agreement of the author or the publisher.

Layout – A S&M QS, Belgrade Copies printed: 1000 Belgrade, September 2005

143 xv Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

PREFACE

The Third International Conference "Total Quality Management - Advanced and Intelligent Approaches" was held from 30th May to 1st June, 2005 at Belgrade. During three days the Conference attended 241 participants from 18th countries.

Selected papers presented at this Conference are now offered to you in the special edition of the TQM&E International Journal. The International Special Edition Editorial Board reviewed the papers and after corresponding additions and corrections they are included in this edition. The papers reflect the world level in theory and practice of different TQM aspects today and as such may be used for education of experts for quality.

The main messages stated during plenary sessions and at the round table are: (i) the support for bi-annual holding of this Conference in Serbia, (ii) presented papers, particularly those from abroad, are a good basis for work on implementation in our country, (iii) the papers from Serbia & Montenegro (the majority) should be in future more oriented towards the results of the implementation, and (iv) the support for national plan of activities for quality improvement in 2005/6.

The Four International Conference with the same global topic and new topics will be held from 27th to 30th May 2007 at Belgrade.

Welcome in Belgrade 2007.

Prof. Dr. V. Majstorović, Conference Chairman

144 xvi Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

Welcome address of Prof. Dr. Vidosav D. Majstorović at the opening of IWC TQM & AIA, 31 May 2005, Belgrade

Distinguished participants, friends from abroad and guests,

I feel great pleasure and honor in being able to welcome you at the Third International Working Conference "TQM 2005". After successful second Conference held two years ago, and praises received from international organizations supporting our work (CIRP, EOQ, EFQM, JUSE, IFIP, IMEKO and IFAC) we find ourselves at the beginning of work of the Third Conference, devoted basically to Total Quality Management, its development and implementation from the aspect of Manufacturing Engineering. In this way we are creating in our country a competent scientific forum for expert analysis of development of system approaches to quality. May I express my great thanks to respected professors and researchers who are co- chairmen of the Conference: Prof. Herbert Osanna from Austria, Mr. Michel Debencham from England, Prof. Albert Weckennman from Germany, Prof. Stanislav Karapetrović from Canada and Prof. Ton Wiele, form the Netherlands. They are prominent and in the world recognized scientists in the fields to be considered at this Conference. Sponsors, important organizations from abroad and from our country, and particularly CIRP, Paris, France and Ministry for Science of Republic Serbia, have significantly contributed to the success of this Conference and I wish to thank them for this. I welcome my professor and colleague, Professor Dr. Vladimir Milačić, from Belgrade University. I welcome all other our respected guests, authors of papers and participants. There is a saying in our country: Welcome and feel at home! and this is what I want to say to you. May I now invite the chairmen of the Opening Plenary Session, Professor Dr. Prof. Dr. Herbert Osanna, and Prof. Dr.-Ing. Albert Weckenmann, Prof. Dr. Ton van der Wiele, Michael Debenham, Prof. Dr. Stanislav Karapetrović, Prof. Dr. Dragutin Stanivuković, and Prof. Dr. Vladimir Milačić, to take their seats at the table. I will be grateful to Professor Dr. Herbert Osanna if he will continue to chair this Session.

Thank you.

Professor Dr. Vidosav MAJSTOROVIC 31 May 2005, Beograd

NOTE: Welcome address of at the opening of IWC TQM & AIA, 31 May 2005, Belgrade are given from following persons: Dr. Branimir Lolić, YUSQ Honorary President; Prof. Dr. Petar Miljanić, Vice - President of Serbian Academy Science and Arts, Belgrade; Vice – president of IMEKO, prof. Dr. Herbert Osanna, Austria; Mr. Michel Debenham, IQA Manager for international cooperation, London, UK; Prof. Dr. Miloš Nedeljković, Dean of Mechanical Engineering Faculty, Belgrade; Mr. Miloš Žugić, Deputy Minister of SCG Gouverment and Prof. Dr. Aleksandar Sedmak, Vice – Minister for Science of Republic Serbia.

145 xvii Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

CONTENTS

01 - INTELLIGENT FLEXIBLE DISASSEMBLY AND RECYCLING OF USED PRODUCTS TO SUPPORT TOTAL QUALITY MANAGEMENT AND SUSTAINABILITY IN EUROPEAN INDUSTRY ¾ Keynote paper P. H. Osanna, M. N. Durakbasa, H. S. Tahirova, Department for Interchangeable Manufacturing and Industrial Metrology, Vienna University of Technology, Wien, Austria 02 - COMPUTER-AIDED DESIGN OF EXPERIMENTS FRAMEWORK - A COMPREHENSIVE APPROACH TO PROCESS IMPROVEMENT ¾ Keynote paper Albert Weckenmann, Peter-Frederik Brenner, Chair Quality Management and Manufacturing Metrology, Erlangen University Erlangen-Nuremberg 03 - A MANAGEMENT CONTROL PERSPECTIVE FOR QUALITY MANAGEMENT: AN EXAMPLE IN THE AUTOMOTIVE SECTOR ¾ Keynote paper Jos van Iwaarden and Ton van der Wiele, Erasmus University Rotterdam, The Netherlands 04 - THE VALUE OF ORGANISATIONAL CULTURE AND THE ROLE OF COMPETENCIES IN DELIVERING QUALITY PRODUCTS AND SERVICES ¾ Keynote paper Michael Debenham, Professional Affairs Manager, Institute of Quality Assurance (IQA), London, United Kingdom. 05 - IMS IN THE M(E)SS WITH CSCS ¾ Keynote paper Professor Stanislav KarapetrovicAuditing and Integration of Management Systems Research Laboratory, Department of Mechanical Engineering, University of Alberta, Edmonton, Canada 06 - A CONTRIBUTION TO THE DIGITAL QUALITY CONCEPT RESEARCH ¾ Keynote paper Professor Vidosav D. Majstorovic, Dr.Sci., Mech. Eng., Mechanical Engineering Faculty, Belgrade, Serbia 07 - ORGANISATIONAL SUSTAINABILITY MANAGEMENT THROUGH MINIMISED BUSINESS EXCELLENCE MODELS ¾ Keynote paper Rickard Garvare1) , Raine Isaksson2), 1) Luleĺ University of Technology, Luleĺ, Sweden, 2) Gotland University, Visby, Sweden. 08 - QUALITY MANAGEMENT MATURITY AND MANAGEMENT ATTITUDE TO QUALITY 1)Mr Milan Ivanović, 2) Prof. Dr. Vidosav D. Majstorović, 1)LR, Belgrade 2)Mechanical Engineering Faculty, Belgrade 09 - DESIGN OF A SYSTEM FOR INTELLIGENT DATA-POINT PRE-PROCESSING IN REVERSE ENGINEERING ¾ Keynote paper I. Budak a), M. Sokovic b), J. Hodolic a), a) Faculty of Technical Sciences, University of Novi Sad, b) Faculty of Mechanical Engineering, University of Ljubljana,Slovenia 10 - „FAST FORWARD” – DESIGNED PROCESS OPTIMISATION Prof. Dr.-Ing. R. Schmitt; Dipl.-Ing. J. Dören, Fraunhofer Institute for Production Technology, Department Metrology and Quality Management; Aachen, Germany 11 - INTERNET BASED AUTOMATION OF THE PROCESS "DESIGNING - MACHINING" OF A WORKPIECE Prof. Dr. K.-D. Bouzakis(1), Assistant Prof. Dr. A. Vakali(2), Lecturer Dr. G. Andreadis(1), E. Karapidakis(2), 1 Laboratory for Machine Tools and Manufacturing Engineering, Mechanical; Eng. Dept., Aristoteles University of Thessaloniki, Greece; 2 Computer Science Dept., Aristoteles University of Thessaloniki, Greece

146 xviii Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

12 - DEVELOPMENT OF AN INTELLIGENT MODULE FOR DECREASE OF MEASURING ERROR ON CMM M. Stevic1, J. Hodolic1 S. Vukmirovic2, 1Insitute for Production Engineering, Faculty of Engineering, University of Novi Sad, Novi Sad, Serbia and Montenegro; 2Department of Control Systems and Automation, Faculty of Engineering, University of Novi Sad, Novi Sad, Serbia and Montenegro 13 - TRACEABILITY OF MEASUREMENT RESULTS IN INDUSTRY Prof. Dr. Bojan. Ačko, University of Maribor, Faculty of Mechanical Engineering, Maribor, Slovenia 14 - SYNERGY IN APPLICATION OF PRACTICES OF THE INTEGRATED PRODUCTION SYSTEM TOWARDS ACHIEVING MANUFACTURING EXCELLENCE ¾ Keynote paper L Sukarma, Agency for the Assesment and Application of Technology, Indonesia 15 - CONTINUOUS IMPROVEMENT – A PREREQUISITE FOR SUCCESS ¾ Keynote paper R. Pendić1, V. Majstorović2, Z. Pendić3, 1 Eurosystems Group – Matheos Invest Group, Belgrade, S&M; 2 Mechanical Engineering Faculty, Belgrade University, S&M; 3 Institute of Nuclear Sciences “VINCA”, Belgrade, S&M 16 - ACCELERATING PERFORMANCE IMPROVEMENT THROUGH APPROPRIATE RESOURCE ALLOCATION L Sukarma, Agancy for the Assessment and Aplication of Technology, Indonesia 17 - APPLICATION OF TAGUCHI MODEL FOR QUALITY PRODUCT IMPROVEMENT Mech. Eng. B.Sc. Tatjana V. Sibalija1, Prof. Dr. Vidosav D. Majstorovich2, 1 STMicroelectronics Malta Ltd., Industry Road, Kirkop, Malta, 2 Mechanical Engineering Faculty, Laboratory for Production Metrology and TQM, Serbia 18 - USING PARTIAL LEAST SQUARES REGRESSION TO ANALYSE QUALITY IN HIGHER EDUCATION ¾ Keynote paper L. Catellani, Bianca M. Colosimo, Q. Semeraro, Dipartimento di Meccanica – Politecnico di Milano, Milano (Italy) 19 - RESEARCH AND DEVELOPMENT OF DIGITAL QUALITY MODEL IN SCM ¾ Keynote paper 1) Nenad Stefanovic, Zastava Automobiles, 2) Vidosav Majstorovic, 3) Dusan Stefanovic, 1)Information Systems Division, Kragujevac, SCG 2)Mechanical Engineering Faculty, Belgrade, SCG; 3)Faculty of Science, Kragujevac, SCG 20 - ISO MANAGEMENT SYSTEMS STANDARDS AND SOCIAL RESPONSIBILITY CONNECTION: (NOT QUITE) JOINED-UP OPINIONS OF ISO’S STAKEHOLDERS Pavel Castka 1), Michaela A. Balzarova 2), 1) University of Canterbury, Christchurch, New Zealand, 2) Brno University of Technology, Brno, the Czech Republic 21 - RESEARCH INTO THE POSSIBILITIES OF THE APPLICATION OF THE FUZZY LOGIC TO THE DEVELOPMENT OF THE QUALITY EVALUATION MODEL Marko Mirkovich, M.Sc.1), Prof. Dr. Dragan Radojevich2) , Prof. Dr. Vidosav D. Majstorovich3), Prof. Dr. Janko Hodolic4), 1 Bauxite mines,s.c. Quality & Information Centre ,Niksic 2 Mihajlo Pupin Institute , Belgrade 3 Mechanical Engineering Faculty , Laboratory for Production Metrology and TQM ,Belgrade 4 FTS, Production Mechanical Engineering Institute, Novi Sad

147 xix Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade

Appendix III :

Fourth International Working Conference FIRST ANNOUNCEMENT, CALL FOR PAPERS, PRESENTATIONS AND PARTICIPATION

TOTAL QUALITY MANAGEMENT – ADVANCED AND INTELLIGENT APPROACHES May 27th – 30th , 2007.Belgrade, SERBIA

International Program Committee (invited): Co – sponsored by (invited): Honorary Chairs: prof. Dr. G. Arndt (Au), M. Japanese Union of Scientits and Engineers Debenham (UK), prof. Dr. S. Karapetrovic (Ca), (JUSE), Tokyo, Japan. prof. Dr. L. Monostori (H), prof. Dr. H. Osanna European Organization for Quality (EOQ), (A), prof. Dr. T. Pfeifer (G), prof. Dr. G. Sohlenius Brusssels, Belgium. (S), prof. Dr. H. VanBrussel (B), prof. Dr. A. European Foundation for Quality Management Weckenmann(G), prof. Dr. A. van der Wiele (Nl). (EFQM), Brussels, Belgium. Conference Chair: Prof. Dr. V. Majstorovic American Society for Quality (ASQ), Milwoku, (S&M). USA Members: prof. Dr. B. Ačko (Sl), prof. Dr. H. Bley International Federation for Information (G), prof. Dr. P. Bertok (Au), prof. Dr. M. Bobrek Processing (IFIP), Laxenburg, Austria. (BiH), prof. Dr. P. Bojanić (S&M), prof. Dr. C. International Measurement Confederation Bouzakis (Gr), prof. D. Brissaud (F), (Miss) L. (IMEKO), Budapest, Hungary. Catellani (I), prof. Dr. Chryssolours (Gr), prof. Dr. International Federation of Automatic Control K. Cho (K), prof. Dr. L. DeChiffre (Dk), B. (IFAC), Laxenburg, Austria. Dimitrijević (Ca), Dr. C. Evans (USA), prof. Dr. G. Halevi (Is), prof. Dr. T. Hardjono (Nl), prof. Dr. S. Organized by Ho (HK), prof. Dr. J. Hodolič (S&M), Dr. R. Mechanical Engineering Faculty, Center for Isaksson (S), MSc. M. Ivanović (S&M), B. Jouslin Advanced Technology, Laboratory for (B), prof. Dr. J. Jedrzejewski (Pl), prof. Dr. Z. Katz Production Metrology and TQM, Belgrade, S&M. (SA), (Mrs) prof. Dr. A. Kjellberg (S), (Mrs) prof. Dr. N. Majkić-Singh (S&M), prof. Dr. V. Milačić Conference Schedule: (S&M), prof. Dr. P. Molnar (H), prof. Dr. G. Morel Sessions based on accepted papers on the (F), prof. Dr. S. Nof (USA), (Mrs) L. Nitu (Ro), S. topics of interest. Olivas (Sp), prof. Dr. V. Parežanin (S&M), Dr. G. Special sessions organized and chaired by Pegs (UK), (Mrs) prof. Dr. S. Pejšić-Tarle (S&M), internationally recognized leaders of research in MSc. Z. Pendić (S&M), (Miss) MSc. T. Sibalija advanced topics. (S&M), F. Steer (UK), prof. Dr. M. Soković (Sl), Tutorials on special topics. prof. Dr. D. Stanivuković (S&M), MSc. N. Stefanović (S&M), prof. Dr. M. Stojčevski (Mk), Conference Venue prof. Dr. R. Teti (I), prof. Dr. J. Rogerson (UK), The Conference will be held at the capital city of MSc. S. Živković (S&M), Dr. V. Yagodzinsky (U), Belgrade, Serbia and Montenegro. prof. Dr. S. Yamada (J), (Mrs) prof. Dr. G. Ušćebrka (S&M). Official Language The official language of the Conference is Sponsored by (invited) English. International Institution for Production Engineering Research (CIRP), Paris, France. Scope of the Conference The main objecive of the Third Conference is to provide an international forum for the exchange

148 xx Final report for CIRP, July 2005 - Prof. Dr. V. Majstorovic Third International Working Conference “ Total Quality Management – Advanced and Intelligent Approaches “ May 30th – June 01st, 2005, Belgrade of knowledge, experience, research results and Important dates information about various aspects of the state- January 15. 2007 – Proposal for special of-the-art and the future development of total sessions and tutorials due. quality management. February 25. 2007 – Full paper due, 8 pages The scope of the Conference covers maximum; electronic submission. philosophical, scientific and practical concepts March 15. 2007 – Final paper acceptance. concerning research, development and March 25. 2007 – Registration announcement. application of TQM-based advanced April 15. 2007 – Final program announcement. approaches. May 15. 2007 – Early registration deadline. May 27 – 30. 2007 – International Working Topics of interest include, but are not limited to: Conference. Business excellence models / BEM(applications and development trends); TQM & manufacturing Conference Program Agenda management; World class performance; First Day – 27. May 2007. / Afternoon – Attractive quality; Robust engineering; Six sigma Industrial tour / Evening – Welcome reception model; Intelligent quality tools and methods; and ROUND TABLE DISSCUSION. Virtual factory and virtual quality; Intelligent Second Day – 28. May 2007. / 10:00a.m. – metrology in manufacturing; Intelligent and noon Opening and Key Note Speakers. Coctail virtual CMM; Business process improvement; / noon - 2:00p.m. Parallel Sessions (1) / 4:00 Breakthrough management; Organizational – 6:00p.m. Panel Discussion – Research Excellence, Intelligent design for quality. Directions / Evening – Informal Conference Intelligent Business. Dinner. Third Day – 29. May 2007. / 10:00a.m. – noon Papers and Proceedings Parallel Sessions (2) / 12.30p.m. – 2:00p.m. Prospective authors are invited to send their Parallel Sessions (3) / 4:00p.m. – contributions on relevant topics. Papers may be 6:00p.m. Panel Discussion – Future submitted by e-mail in a Word file . The length Engineering Education for TQM. of the paper should be 8A4 pages, 1.0 space Fourth Day – 30. May 2007. / 10:00a.m. - typing, use arial 10 point font. For the noon Closing Plenary Session. anonymity of review, please identify the following information separately: the title, Correspondence – Conference Secretariat: author's full name, affiliation, address, e-mail Prof. Dr. Vidosav D. MAJSTOROVIĆ address, telephone and fax numbers. Authors of Mechanical Engineering Faculty accepted papers will be expected to sign a Kraljice Marije 16 copyright release allowing publication in the 11000 Beograd proceedings. Pre - proceedings will be available Serbia before the Conference. After the Conference, the Proceedings with selected papers which Phone: ++ 381 (0)11 30 70 407 Fax: ++ 381 will under a second review, RTD, panel (0) 33 70 346. discussion and conclusion will be published in the International Journal “ Total Quality E - mail: [email protected] Web site: Management and Excellence”. www.mas.bg.ac.yu or www.iwc.tqm.yu

Registration The registration fee for the Conference is EURO 150, if paid before 15th May 2007. After that date, the fee is EURO 200. For the registration fee please contact the Conference Secretariat. Report prepared by Prof. Dr. Vidosav D. MAJSTOROVIC, Dipl. Lodging Mech. Eng. For details regarding hotel reservations please IWC TQM Conference Chairman and Founder contact the Conference Secretariat. In Belgrade, 15th July, 2005.

149 xxi

Mechanical Engineering Faculty, Center for Advanced Technology, Laboratory for Production Metrology and TQM, Kraljice Marije 16 , Belgrade CIRP - International Institution for Production Engineering Research, Paris, France YUSQ - Association Serbia & Montenegro for Quality and Standards, Kneza Milosa 9/II, Belgrade

Fourth International Working Conference FIRST ANNOUNCEMENT, CALL FOR PAPERS, PRESENTATIONS AND PARTICIPATION TOTAL QUALITY MANAGEMENT - ADVANCED AND INTELLIGENT APPROACHES

Correspondence - Conference Secretariat: Prof. Dr. Vidosav D. MAJSTOROVIC Mechanical Engineering Faculty , The Panorama of FME Belgrade Kraljice Marije 16, 11000 Belgrade, Serbia Phone: ++ 381 (0)11 30 70 407 Fax: ++ 381 (0)11 33 70 346 May 27 - 30, 2007., Belgrade , SERBIA & MONTENEGRO E-mail: [email protected] Web site: www.mas.bg.ac.yu www.iwc.tqm.yu www.jusk.org.yu ZZZLZFWTP\X ,QWHUQDWLRQDO3URJUDP&RPPLWWHH &RQIHUHQFH6FKHGXOH ,PSRUWDQWGDWHV,PSRUWDQW GDWHV +RQRUDU\&KDLUV Sessions based on accepted papers on the topics of interest. January 15. 2007 - Proposal for special sessions and prof. Dr. G. Arndt (AU), M. Debenham (UK), prof. Dr. S. Special sessions organized and chaired by internationally recognized tutorials due. Karapetrovic (Ca), prof. Dr. L. Monostori (H), prof. Dr. H. Osanna leaders of research in advanced topics. February 25. 2007 - Full paper due, 8 pages maximum; (A), prof. Dr. T. Pfeifer (G), prof. Dr. G. Sohlenius (S), prof. Dr. H. Tutorials on special topics. electronic submission. VanBrussel (B), prof. Dr. A. Weckenmann(G), prof. Dr. A. van March 15. 2007 - Final paper acceptance. der Wiele (Nl). &RQIHUHQFH9HQXH March 25. 2007 - Registration announcement. The Conference will be held at the capital city of Belgrade, Serbia and April 15. 2007 - Final program announcement. &RQIHUHQFH&KDLU Montenegro May 15. 2007 - Early registration deadline. Prof. Dr. V. Majstorovic (S&M). 0D\- International Working Conference. 2IILFLDO/DQJXDJH 0HPEHUV The official language of the Conference is English. &RQIHUHQFH3URJUDP$JHQGD prof. Dr. B. Ačko (Sl), prof. Dr. H. Bley (G), prof. Dr. P. Bertok &FirstRQ IDayHUH Q- F27.H 3MayURJ U2007.DP $ J/ HAfternoonQGD - Industrial tour / (Au), prof. Dr. M. Bobrek (BiH), prof. Dr. P. Bojanić (S&M), prof. 6FRSHRIWKH&RQIHUHQFH Evening - Welcome reception and ROUND TABLE Dr. C. Bouzakis (Gr), prof. D. Brissaud (F), (Miss) L. Catellani (I), The main objecive of the Third Conference is to provide an DISSCUSION. prof. Dr. Chryssolours (Gr), prof. Dr. K. Cho (K), prof. Dr. L. international forum for the exchange of knowledge, experience, Second Day - 28. May 2007. / 10:00a.m. - noon DeChiffre (Dk), B. Dimitrijević (Ca), Dr. C. Evans (USA), prof. Dr. research results and information about various aspects of the state-of- Opening and Key Note Speakers. Coctail / noon - G. Halevi (Is), prof. Dr. T. Hardjono (Nl), prof. Dr. S. Ho (HK), prof. the-art and the future development of total quality management. 2:00p.m. Parallel Sessions (1) / 4:00 - 6:00p.m. Panel Dr. J. Hodolič (S&M), Dr. R. Isaksson (S), MSc. M. Ivanović The scope of the Conference covers philosophical, scientific and Discussion - Research Directions / Evening - Informal (S&M), B. Jouslin (B), prof. Dr. J. Jedrzejewski (Pl), prof. Dr. Z. practical concepts concerning research, development and application Conference Dinner. Katz (SA), (Mrs) prof. Dr. A. Kjellberg (S), (Mrs) prof. Dr. N. of TQM-based advanced approaches Third Day - 29. May 2007. / 10:00a.m. - noon Parallel Majkić-Singh (S&M), prof. Dr. V. Milačić (S&M), prof. Dr. P. Sessions (2) / 12.30p.m. - 2:00p.m. Parallel Sessions (3) Molnar (H), prof. Dr. G. Morel (F), prof. Dr. S. Nof (USA), (Mrs) L. 7RSLFVRILQWHUHVWinclude, but are not limited to: / 4:00p.m. - 6:00p.m. Panel Discussion - Engineering Nitu (Ro), S. Olivas (Sp), prof. Dr. V. Parežanin (S&M), Dr. G. Business excellence models / BEM(applications and development Education for TQM. Pegs (UK), (Mrs) prof. Dr. S. Pejšić-Tarle (S&M), MSc. Z. Pendić trends); TQM & manufacturing management; World class Fourth Day - 30. May 2007. / 10:00a.m. - noon Closing (S&M), (Miss) MSc. T. Sibalija (S&M), F. Steer (UK), prof. Dr. M. performance; Attractive quality; Robust engineering; Six sigma model; Plenary Session. Soković (Sl), prof. Dr. D. Stanivuković (S&M), MSc. N. Intelligent quality tools and methods; Virtual factory and virtual quality; Stefanović (S&M), prof. Dr. M. Stojčevski (Mk), prof. Dr. R. Teti Intelligent metrology in manufacturing; Intelligent and virtual CMM; (I), prof. Dr. J. Rogerson (UK), MSc. S. Živković (S&M), Dr. V. Business process improvement; Breakthrough management; Yagodzinsky (U), prof. Dr. S. Yamada (J), (Mrs) prof. Dr. G. Organizational Excellence, Intelligent design for quality. Intelligent Ušćebrka (S&M). Business.

6SRQVRUHGE\ 3DSHUVDQG3URFHHGLQJV International Institution for Production Engineering Research ( Prospective authors are invited to send their contributions on relevant CIRP ) , Paris, France. topics. Papers may be submitted by e-mail in a Word file . The length of the paper should be 8A4 pages, 1.0 space typing, use arial 10 &RVSRQVRUHGE\ point font. For the anonymity of review, please identify the following Japanese Union of Scientits and Engineers (JUSE), Tokyo, information separately: the title, author's full name, affiliation, address, Japan. e-mail address, telephone and fax numbers. Authors of accepted European Organization for Quality (EOQ), Brusssels, Belgium. papers will be expected to sign a copyright release allowing European Foundation for Quality Management (EFQM), publication in the proceedings. Pre - proceedings will be available Brussels, Belgium. before the Conference. After the Conference, the Proceedings with American Society for Quality (ASQ), Milwoku, USA selected papers which will under a second review, RTD, panel International Federation for Information Processing (IFIP), discussion and conclusion will be published in the International Journal Laxenburg, Austria. " Total Quality Management and Excellence". . International Measurement Confederation (IMEKO), Budapest, Hungary. 5HJLVWUDWLRQ International Federation of Automatic Control (IFAC), The registration fee for the Conference is EURO 150, if paid before Laxenburg, Austria. 15th May 2007. After that date, the fee is EURO 200. For the registration fee please contact the Conference Secretariat 2UJDQL]HGE\ Mechanical Engineering Faculty, Center for Advanced Technology, Laboratory for Production Metrology and TQM, /RGJLQJ Belgrade, S&M. For details regarding hotel reservations please contact the Conference Secretariat &RRUJDQL]HGE\ Association Serbia & Montenegro for Quality and Standards, Belgrade, S&M

YUSQ -QU ALIT Y ANDS T AND ARDIZA TION ASSOCIA TION OFSERBIA &MONTENEGR O

UDK 658.5 YU ISSN 1452-0680 No 3 Vol. 33 2005 Management fr P I n a om 30thMa t p e MEN YUSQ -QU e r MAN n r s T a

t ADžMENT T i s o ALIT e O n EX c a o Y ANDS '' Specialedition l n

y t - T d A

W o 1s dv T r AL QU AND e CELLENCE o INTERN A anced v O r ARDIZA t June,2005,atBelgr i k T e ALNIM KV i w n GEMENT & g e TION ASSOCIA A

d and TION

C a o n AL JOURN Int n d ALITET f

e elligent TION OFSERBIA&MONTENEGR p r e r ALIT e n s c AL OM &IZVRSNOS e e n

t Appr e ade, Ser ' d ' T

o a oaches'', t t a

l Y

t h bia. e Q O

u T T a held h l i i t r y d