UDK 658.5 YU ISSN 1 452-0680 N o 3 V ol. 33 2005
INTERNATIONAL JOURNAL
'' 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.
TOTAL QUALITY MANAGEMENT &
EXCELLENCE
MENADžMENT TOTALNIM KVALITETOM & IZVRSNOST
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.
ii
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
University of Wollongong, Wollongong, Australia
Prof. dr Daniel BRISAUND
University of Grenoble, Grenoble, France Prof. dr Barrie DALE
UMIST, Manchester, England Michel DEBENHAM
Institute of QA, London, England Sofija DJURDJEVIC
YUQS, Belgrade, S&M Prof. dr Noriaki KANO
Science University of Tokyo, Tokyo, Japan Prof. Dr. Laslo MONOSTORI
Hungarian Academy, Budapest, Hungary Prof. dr Vidomir PAREŽANIN
ASTRA grupa, Beograd Mr Zoran PENDIĆ Institut Vinča, Beograd
Prof. dr Gunnar SOHLENIUS
Royal Institute of Technology, Stockholm, Sweden
Prof. dr Dragutin STANIVUKOVIĆ FTN, Novi Sad, S&M
Prof. dr Herbert OSANNA
Technical University, Wien, Austria Prof. dr Tilo PFIEFER
RWTH Aachen, Aachen, Germany Prof. dr Hendrik VanBRUSSEL
Katolike University Leuven, Heverlez, Belgium Prof. dr Shu YAMADA
Science University of Tokyo, Tokyo, Japan Prof. dr Albert WECKENMANN
University Erlangen, Erlangen, Germany Prof. dr Ton ven der WIELE
Erasmus University Roterdam, The Netherlands
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: jusk@EUnet.yu 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.
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
iv
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 Garvare
1), Raine Isaksson
2),
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
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),
1Laboratory for Machine Tools and Manufacturing Engineering, Mechanical;
Eng. Dept., Aristoteles University of Thessaloniki, Greece;
2Computer Science Dept., Aristoteles University of Thessaloniki, Greece
12 - DEVELOPMENT OF AN INTELLIGENT MODULE FOR DECREASE OF MEASURING ERROR ON CMM………..………. 61 M. Stevic
1, J. Hodolic
1S. Vukmirovic
2,
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. Pendić
1, V. Majstorović
2, Z. Pendić
3,
1Eurosystems Group – Matheos Invest Group, Belgrade, S&M;
2Mechanical Engineering Faculty, Belgrade University, S&M;
3Institute 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. Sibalija
1, Prof. Dr. Vidosav D. Majstorovich
2,
1STMicroelectronics Malta Ltd., Industry Road, Kirkop, Malta,
2Mechanical 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
vi
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 Radojevich
2), Prof. Dr. Vidosav D. Majstorovich
3), Prof. Dr. Janko Hodolic
4),
1Bauxite mines,s.c. Quality & Information Centre ,Niksic
2Mihajlo Pupin Institute , Belgrade
3Mechanical Engineering Faculty , Laboratory for Production Metrology and TQM ,Belgrade
4FTS, 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
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
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
In modern industrial production the very important issues of the protection of environment and sustainability must be taken into consideration increasingly /1, 2/.
Manufacturers of products for every day life - consumer electronics, automobiles, house hold devices - are facing increasing requirements of consumers, public opinion and governments to minimize the pressure on environment - increasing consumption of raw materials and energy, growing pollution and waste.
The complexity of scrap recycling from the above mentioned consumer products is a strongly interdisciplinary field where there are items like collection logistics, disassembly, components re-use, recovery of precious and rare materials like copper, recycling of non-hazardous materials and disposal of hazardous and toxic substances. All these require the coordinated work of interdisciplinary teams.
European Union Environment Council approved in 1990 the Commission’s Strategy for Waste Management, which included ELVs (End of Life Vehicle) as a Priority Waste Stream.
Concerning of the Waste Strategy Priorities the generally accepted hierarchy of priority with respect to waste is:
• prevention,
• recovery,
• disposal.
Within the term recovery the concept includes the reuse of parts, the recycling of material and the recovery of energy.
The Objectives of the European Commission Proposal for Automobiles are the following:
- Avoidance of waste,
- Reduction of landfill demand, and - Reduction of toxicity.
European “Take-Back Law” requires automobile manufacturers to take back all vehicles which were ever sold in that country. Voluntary agreements have been widely accepted by industry and the threat of legislation has subsided slightly.
Essential advantages to application of old automobiles are established by the following directives of EU and according to their transformation to the national right:
•
The directive 75/442/ European Economic Community of advice from July 15, 1975, about waste.•
The directive 96/61/ European Economic Community of advice from September 24, 1996, about the integrated evasion and reduction of environmental contamination (IPPC - the instruction).•
The directive 1999/31/ European Economic Community of advice from April 26, 1999, about gathering waste.•
The directive in 2000/76/ European Economic Community of the European parliament and advice from December 4, 2000, about combustion of waste.•
The directive in 2000 / 53 / European Economic Community of the European parliament and advice from September 18, 2000, about old automobiles.The overall targets for the improved process, which are in line with those proposed by the EU, are:
• for a car being scrapped in 2002, a maximum of 15% of initial weight to go to landfill
• for a car being scrapped in 2015, a maximum of 5%
of initial weight to go to landfill
Economic viability is a necessity for the implementation of the improved disposal process under market conditions. If analysis of detailed monitoring information indicates that an action is viable, but direct financial support is required for it to be sustainable, then the means of providing such support will be considered.
Keynote paper
Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005
Currently, the European Commission is proposing a 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 responsibility between all affected parties.
Essential targets in accordance with item 7 of the EU instructions 2000/53/EG about old vehicle are /3/:
•
Target from 2006:85 % recycling
80 % material recycling
that is a maximum of 15 % disposal.
•
Target from 2015:95 % recycling
85 % material recycling
that is a maximum of 5 % disposal.
2 LIFE CYCLE COSTS
In the past the cost structure of a product contained only cost of development, production including quality assurance, marketing, sales and service. As this is mainly influenced by the manufacture of the product it is important to take into consideration the Life Cycle Costs what means that the real costs in each life stage are added (see Figure 1).
Figure 1: Cost Influences During Product Life Cycle
For this purpose a detailed quantitative assessment of expenses for a product has to be carried out taking into account the following:
- Development costs:
Roughly 5 % of the life cycle costs originate from this stage. On the other side to this small part of the costs between 60 % up to more than 80 % of the total life cycle costs are fixed during the development stage. The influences of other stages on the cost level are small compared with those of research and development.
- Production costs:
In any cases one can operate on the rough rule which allows the determination of production costs based on costs of different materials used for engineering purposes
- Use costs:
The costs of the product use phase can be calculated on the basis of the energy costs per time unit that the product is in use.
- Disposal costs:
The costs of take-back, recycling and final disposal are comprised by the costs of take-back systems, recycling and final disposal processes as well as the returns for the gained secondary raw materials.
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.
3 END OF LIFE MANAGEMENT
The challenges for innovative product design are two fold: on the one side having to contain material and energy flows within the product life cycle to close the loop, on the other side reducing the overall consumption of materials and energy to make the loop thinner (see Figure 2). These both should be achieved whilst at the same time the consumer needs are to be identified and met. The closed life cycle consists therefore of two main streams:
- The "traditional" product life chain (such as production, distribution and use). Introducing environmental concerns in this stream is called "Eco-design".
- Re-using and recycling products, components and materials (such as take back, re-use and recycling).
Operation of this stream is called "End-of-life Management".
By disassembly of further on usable parts considerable reduction of use costs and disposal costs can be achieved.
In this respect it is very important that different branches of industry collaborate insofar that different experiences are exchanged.
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 System
A fixed automated single purpose disassembly system cannot operate economically at the time being. So it is very important to design such a system with high as possible flexibility. Such a system can be called
"Intelligent Flexible Disassembly Cell" (see Figure 3) and it consists of different main modules:
- Disassembly Robot or handling device with special features like path and force control
- Robot Gripper for a wide spectrum of parts with different geometry and dimensions
- Disassembly Tools especially developed for robots - Components Data Base including data of re-usable and re-manufacturable parts
- Storage Device for tools and parts
- Transport System and feeding system for products to be disassembled
- Clamping Device and fixture system for parts with different geometry and dimensions
- Manual Disassembly Station for specific tasks - Sensors for force, torque, visual recognition, position and distance
- Intelligent Cell Control Unit able to process information from extended sensors.
Figure 3: Modules of an Intelligent Flexible Disassembly Cell.
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
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
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 to all members of the organization an to society /5]. One key element of TQM is to use quality management tools and techniques for continuous process improvement /2].
However, most quality management tools and techniques 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 technique for continuous improvement of manufacturing processes is Design of Experiments (DoE) /1], /3], /6].
The application of DoE requires trained and skilled employees. To improve the application of complex quality management techniques, it is necessary to develop computer-aided tools to support the user. By means of computer-aided DoE tools, employees can strengthen their methodical skills in DoE and the user is assisted in designing experiments.
Definition of Design of Experiments
Design of Experiments is mainly used during process planning, manufacturing and assembly.
In these phases of the product life cycle, a model of the existing processes is usually used to explain why Design of Experiments should be applied. Assignable causes (inputs) and extraneous variables (stochastic inputs) affect the response variables (outputs). The cause- and-effect relationships within the process are partially or completely unknown (Fig. 1).
In Design of Experiments, inputs are understood to be assignable causes that affect the response variable. The response variables are outputs that describe the observed result of experiments. An unintentional and uncontrollable influencing variable is termed an extra- neous variable. The influencing variables used for the experiments are also called factors.
Figure 1: Process model
In literature, Design of Experiments is described as follows:
• /ISO 3534-3]
“The arrangement in which an experimental program is to be conducted, and the selection of the levels (versions) of one or more factors or factor combinations to be included in the experiment."
• For an accurate definition, refer to /4]
“A designed experiment is a test in which some purposeful changes are made to the input variables of a process or system so that we may observe and identify the reasons for change in the output response."
The procedure followed in DoE differs from a series of experiments as they are conducted in practice.
Normally, one influencing parameter is changed and the effect on the response variables is observed. The change to the influencing variables is therefore somewhat intuitive. If the required result is not achieved by changing one parameter or if a further improvement is expected with a certain setting, another influencing variable is varied. The results are optimum response variables obtained largely by chance.
The path there is no longer reproducible. It is not known which influencing variable really caused the
extraneous variables
Black Box process model
inputs output
s Keynote paper
Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005
improvement in the response variables. In design of experiments, an attempt is made to change several influencing variables at once with a systematic procedure and to represent their effect on the response variables in a reproducible way.
The goals of DoE are as follows:
• to determine assignable causes that have a significant effect on response variables,
• to obtain information about the project, process, and machine,
• to optimize the quality of the products and processes,
• to conduct acceptance tests on machines, work pieces and processes.
Methodical approach of Design of Experiments The methodical approach in Design of Experiments can be divided into five phases: system analysis, experimental strategy, realization of experiments, evaluation and validation. For all of these phases, specific DoE knowledge must be provided to gain new information about the process or the product (Fig. 2).
Figure 2: Phases of Design of Experiments The aim of the phase system analysis consists of
discovering process or product variables, called factors, which influence the process yield output or the product specifications. The identification of the factors is done with the help of systematic quality management tools like Ishikawa-Diagrams, Brainstorming and Metaplan- Technique.
The phase experimental strategy deals with the selection of an appropriate experimental design. For this, the DoE user must have a deeper knowledge of the characteristics of the different experimental designs like full factorial or fractional factorial designs. Furthermore, relevant factors with a supposed impact on the process output or the product quality have to be chosen and levels have to be defined.
During the phase realization of experiments, the planned experiments are conducted /4]. Within this phase, the factors have to be adjusted with respect to the determined levels and the experiments have to be run in a careful way.
The phase evaluation deals with the mathematical analysis of the experimental results. The focus lies on the application of mathematical and statistical methods as well as the graphical analysis of the effects. The objective is to investigate optimum settings of the factors as process parameters for achieving the optimized output.
In the phase validation, confirmation experiments have to be run in order to confirm if the proposed factor settings are right. This includes the documentation of the new information concerning the product or the process.
DoE knowledge Process information
- Ishikawa diagram - Brainstorming
- factors
- noise variables - interactions
- experimental data - full factorial designs
- fractional factorial designs - central composite designs
- calculation of effects - analysis of variance - regression analysis - determination of the confidence interval
- significant factors and interactions - mathematical models
- optimal settings of factors
- possibilities of extension of the experiments
System analysis
Evaluation
Realisation of experiments Experimental strategy
Validation
- randomization6
Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005
Without the mentioned DoE knowledge it is not possible for a worker or an engineer, responsible for a process or a machine, to plan, run and evaluate experiments in a meaningful way. For the user it is very important to know how to select the proper DoE methods to conduct meaningful and economic experiments as well as to apply the selected methods in very short time on the process. For this, it is necessary to develop user-friendly tools which guide and assist the DoE user directly and
“just in time” in all phases of DoE /7]. To supply the user with the DoE knowledge “just in time”, the tools should be computer-aided.
Requirements of a computer-aided Design of Experiments framework
Regarding the above mentioned phases of DoE and the necessary knowledge, the following requirements of a computer-aided DoE framework can be pointed out:
• to apply the different techniques, methods and tools of DoE in an efficient way, the user should have direct access to the DoE knowledge where it is needed.
Furthermore, the user should be able to read up on the fundamental quality tools like the Ishikawa-Diagram in order to use these tools for the system analysis.
• process information obtained during the phase system analysis regarding the process or the product like factors, interactions and noise variables should be documented consistently. Based on that primary process information, cause and effect relationships between the single steps of a process chain can be evaluated. In addition the system should contain learning algorithms which help the user when drafting experimental strategies
based on the already known process and product information.
• the user should be advised when choosing adequate experimental designs which fulfill the objective of the planned experiments as well as in selecting of mathematical and statistical methods. Furthermore the user should be advised in the application of adequate methods and tools available with quality management.
• to guide the user through the different phases of DoE and to assist him in regard to his specific tasks when planning experiments, so called assistants should be included in the system. These assistants should help the user in planning experiments without having a big theoretical background in statistical tools.
In general the compatibility of the system with existing software packages for DoE which are especially designed for evaluating raw data of the experiments is required. That means that the data gained from the system analysis phase and raw data from the experiments should be transferred to statistical software packages via an interface so that the analysis of the experimental results can easily be managed. Furthermore the computer-aided DoE framework must be capable to be integrated into an existing computer environment like CAx tools.
Concept of the computer-aided Design of Experiments framework
To provide the user with access to DoE knowledge i.e. directly on the shop floor, the computer-aided Design of Experiments framework is designed as a web-based system. The user can access the system via a standard web-browser installed on a standard personal computer (Fig. 3).
Figure 3: Concept of the computer-aided Design of Experiments framework
The main components of the computer-aided Design
of Experiments framework are: • Hypertext system: the features of a hypertext system are nodes and links. Nodes can contain text, graphics, audio, video, animation and images while links Beratung
des Benutzers
Verweis auf Hypertextsystem
data-base for process information
access to assistants input output
integration in hypertext documents
assistants user interface in
web-browser
W E B S E R V E R knowledge-based
system
hypertext system
for DoE knowledge
integration of process information direct access
to DoE knowledge
linking to hypertext system advisor
Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005
connect nodes related in a certain manner. The linking capability allows the non-linear organization of information in the hypertext system. In the hypertext system, the explicit knowledge concerning the different techniques, methods and tools for DoE is stored according to the five phases of DoE. The usage of a hypertext system allows the user to navigate through DoE knowledge on demand and just in time, via the web-based interface of the system.
• Knowledge-based system: the knowledge-based system contains the formulated knowledge concerning the different types of experimental designs like full factorial, fractional factorial and central composite designs. Via a web-based question-answer-dialogue the user is guided through a rule-based selection of adequate experimental designs step-by-step. For this, the user is asked to specify the number of factors and the number of factor levels as well as pointing out the objectives of the planned experiments i.e. running screening experiments or running optimizing experiments. The knowledge-based system explains with the help of the hypertext system the recommended experimental designs. Furthermore the 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 system the user can store the results of the system analysis phase as the initial step of DoE as well as the results of the experiments. The web-based access to this data base allows the user to get relevant process information directly on the shop floor. Significant factors influencing the yield output and optimal process parameter settings obtained by the planned experiments are documented consistently.
• 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 designing experiments without having a strong statistical background. An assistant is implemented in the system, which helps the user to calculate the required sample size to conduct experiments with high statistical power.
The proposed framework was realized for a process chain in electronics manufacturing. The benefits of the framework are:
• employees such as engineers are trained directly where they apply the DoE knowledge, they have just in time and on demand access to the hypertext modules,
• the DoE framework guides the user through the different phases of DoE,
• data from the process chain is collected, structured and analyzed.
CONCLUSIONS
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.
The data-base system allows the user to collect, to structure and to analyze process data and process information which is needed for continuous improvement. User guidance and user assistance during the phases of Design of Experiments (DoE) is very important for the meaningful application of experimental methods for continuous process improvement. Because of the minimizing of processing cycles in distributed process chains, the factor time becomes a critical success factor for production systems and therefore for the application of DoE within complex process chains.
To optimize the yield output of process chains, the DoE user must apply the different techniques, methods and tools in very short time. For this a computer-aided DoE framework was developed. The system 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. In forthcoming work, the system will be extended with learning methods and the results will be transferred to the industry.
ACKNOWLEDGEMENTS
The current work represents results of the research project “Design of Experiments for Manufacturing Processes of finely structured Electronic Products with Numerous Variations” of Collaborative Research Centre SFB 356 "Production Systems for Electronics" funded by the Deutsche Forschungsgemeinschaft, DFG (German Research Foundation).
REFERENCES
/1] Anderson, M. J.; Whitcomb, P. J.: DoE simplified.
New York: Productivity Press Inc., 2000
/2] Dalken, B. G.; Cooper, C. L.; Wilkinson, A.:
Managing Quality and Human Resources. Oxford: Blackwell Publishers Inc., 1997
/3] Fasser, Y.; Brettner, D.: Process improvement in electronics production. New York: Wiley, 2003
/4] Montgomery, D.C.: Design and Analysis of experiments. New York: Wiley, 1991
/5] Pfeifer, T.: Quality Management: Strategies, Methods, Techniques. München: Hanser Verlag, 2002
/6] Scheffler, E.: Statistische Versuchsplanung und - auswertung: Eine Einführung für Praktiker. Stuttgart: Deutscher Verlag für Grundstoffindustrie, 1997
/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 Innovations ISI'2001 (Congress IQMM 2001) Dubai, U.A.E 17- 21. March 2001). 2001, pp. 983-989.
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
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
The impact of the two trends of increasing product variety and shortening product life cycles on organizations lies in their ability to increase complexity and uncertainty. If companies have to produce a growing number of different products and on top of that have to constantly introduce new versions of all these products, they are operating in a complex and unpredictable environment. The complexity is caused by the large number of different processes that all require
management attention. It is clearly more straightforward to manage a single mass production process than to manage a number of production processes with large product varieties. The unpredictability is caused by the constant flow of new product introductions and product updates. Because they imply that success in the market place may last only shortly (until your competitors introduce new versions of their products). Any business environment is a mixture of stability (predictability) and instability (adaptation to changes) (Prater et al., 2001).
Yet increasing product variety and shortening product life cycles are moving many firms towards more
Keynote paper
Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005
unpredictability and instability. A survey among quality experts led to the conclusion that quality management has 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 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 distinguish between, on the one hand, simple and stable environments and, on the other hand, complex and unpredictable environments. Existing quality models like the quality award and business excellence models (e.g.
Malcolm Baldrige National Quality Award, European Business Excellence Award, and the Deming Prize) are not appropriate for this purpose because they do not make this distinction. Therefore, it is needed to look for an appropriate model outside the quality field.
On the basis of a literature review it can be concluded that quality management consists of three core building blocks, which are (1) customer orientation, (2) process control, and (3) continuous improvement (Dean and Bowen, 1994; Wilkinson et al. 1998; Handfield and Melnyk, 1998; Dale et al., 2000; Dale, 2003). Based on these three building blocks it can be argued that quality management can be seen as a control system because all three building blocks aims to control an organization’s processes and to improve and change these processes in response to changes. Therefore, a logical place to look for a model is in the field of management control.
MANAGEMENT CONTROL
A model in the field of management control that can distinguish between, on the one hand, simple and stable environments and, on the other hand, complex and unpredictable environments, is Simons’ four levers of control model (Simons, 1995). This model is shown in figure 1. Simons’ four levers of control model is used to balance control mechanisms in an organization in order to realize the business strategy. The model distinguishes four different types of control mechanisms: (1) beliefs systems, (2) boundary systems, (3) diagnostic control systems, and (4) interactive controls systems. Two of these four levers increase individual freedom (i.e. beliefs systems and interactive control systems), and two restrict individual freedom (i.e. boundary systems and diagnostic control systems).
Business Strategy Core
Values
Risks to be Avoided
Strategic Uncertainties
Critical Performance
Variables Interactive
Control Systems
Diagnostic Control Systems Beliefs
Systems
Boundary Systems
Figure 1: Simons’ four levers of control model (Simons, 1995) Beliefs systems
Beliefs systems are used to inspire and direct the search for new opportunities. A beliefs system is the explicit set of organizational definitions that senior managers communicate formally and reinforce systematically to provide basic values, purpose, and direction for the organization. The definitions espouse the 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 credos, mission statements, and statements of purpose.
Boundary systems
Boundary systems are used to set limits on opportunity-seeking behaviour. Boundary systems delineate the acceptable domain of activity for organizational participants. Unlike beliefs systems, boundary systems do not specify positive ideals. Instead, they establish limits, based on defined business risks, to opportunity seeking.
Diagnostic control systems
Diagnostic control systems are used to motivate, monitor, and reward achievement of specified goals.
Diagnostic control systems are the formal information systems that managers use to monitor organizational outcomes and correct deviations from preset standards of performance. These feedback systems, which are the 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 to activate search. Interactive control systems focus attention and force dialogue throughout the organization.
They provide frameworks, or agendas, for debate, and motivate information gathering outside of routine channels. These control systems stimulate search and learning, allowing new strategies to emerge as participants throughout the organization respond to perceived opportunities and threats. An interactive control system is not a unique type of control system:
many types of control systems can be used interactively by senior managers.
The four different control levers in the model of Simons and their relation to strategy are summarized in table 1.
Table 1: Relating the four levers of control to strategy
Control
system Purpose Communicates Control of strategy as
Beliefs systems
Empower and expand search activity
Vision Perspective
Boundary systems
Provide limits of freedom
Strategic
domain Competitive position
Diagnosti c control systems
Coordinate and monitor the implementa tion of intended strategies
Plans and goals Plan
Interactiv e control systems
Stimulate and guide emergent strategies
Strategic
uncertainties Pattern of actions
Source: Simons, 1996, p. 304
Any control system in an organization can be classified according to the types that Simons distinguishes. The four different types of control systems work together to realize the business strategy. To be able to do this successfully, there should be a balance between the different types of control systems. If there is too much focus on just one or two types of control systems, the organization may have difficulties in realizing its 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 operating. If the environment is predictable and not complex, an organization can put more emphasis on the diagnostic control systems and boundary systems.
However, if the environment is unpredictable and complex, a stronger focus on beliefs systems and interactive control systems is necessary. In the current environment the two trends of increasing product variety and shortening product life cycles bring about unpredictability and complexity for many organizations.
The same reasoning that applies to general control systems also applies to specific control systems, like quality management. Therefore, quality management systems can also be classified according to the four levers of Simons’ model. Moreover, the two mentioned trends of increasing product variety and shortening product life cycles are expected to have a major impact on quality management. So, a shift is expected from a major focus on diagnostic quality management systems to a more important role for interactive quality management systems.
METHODOLOGY
To empirically test the hypothesized shift towards an increased importance of beliefs systems and interactive control systems in a situation of uncertainty and complexity, three case studies have been conducted at European automotive manufacturers (OEMs). Because of the explorative nature of this research, a case study approach is a suitable methodology (Yin, 2003).
In both organizations interviews have been held with the quality manager, supply chain manager, logistics manager, production manager, and human resources manager. On top of that three relevant first tier suppliers have been selected for each of the OEMs. At each of these suppliers interviews were held with the account manager for the OEM (in some cases together with the quality manager of the supplier).
All interviews were conducted by two interviewers and each interview took between 1.5 and 3 hours. The interviews inside the OEM organization were all focused at the changes in management systems that have taken place over the last ten years in the field of responsibility of the interviewee. The interviews at the suppliers focused on the changes that have taken place in the way the relationship between the supplier and the OEM is managed by the OEM.
Each interview was written down by both interviewers and, based on these two write-ups, a final write-up of the interview was produced. Out of these write-ups the most important quality management issues and developments were derived by means of discussions about the write-ups by a group of academic quality experts. The derived issues and developments were presented to the interviewees during a discussion meeting in which these managers could express their perceptions and opinions.
RESULTS
The results of the case studies at the three OEMs have been interpreted by means of the Simons model by placing the identified quality management issues and developments in the four levers. This means that each issue or development has been judged by a group of academic quality experts and positioned in one or more levers of the Simons model. Table 2 shows an example of
Special Edition “Total Quality Management & Excellence” No.3 , Vol. 33 , 2005
developments and issues in the area of process control in a European automotive manufacturer. The numbering indicates the sequence of developments. Solid arrows indicate a development that has happened already, while
dashed arrows indicate a current development or planned/expected future development for which already some evidence has been found.
Table 2: An example of developments and issues in an OEM’s process control
Beliefs
systems Interactive control systems Boundary
systems Diagnostic control 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 years the importance of the supply chain has increased for this manufacturer. Parallel to that there has been an evolution in the management and control of the quality of the suppliers. More than five years ago this manufacturer used incoming inspection to diagnostically measure the quality of the products that were delivered by its suppliers. So, in this stage the quality management system used was a diagnostic control system because it measured the compliance to preset quality standards.
When this manufacturer realized how time and resource consuming this policy was, it started to move towards quality assurance by means of supplier assessments and ratings. The quality performance of each supplier is monitored on a day-to-day basis and recorded in a supplier database. The ratings of suppliers are used in the decision process that takes place when new supply contracts will be given to suppliers. This is again a quality management system that is diagnostic in nature because it uses predetermined performance measures that lead to a ranking of suppliers from which the top performing suppliers will be selected for future contracts.
In recent years, a development has been started towards co-development between this manufacturer and its suppliers. This entails that there is communication and discussion between the manufacturer and its suppliers about new products. So, the manufacturer shows its plans for future products to its suppliers and it asks these suppliers to comment on the plans in order to tackle possible quality problems while the product is still in the design phase. In an interactive way, the manufacturer discusses its new designs with suppliers and uses the
knowledge and experience of these suppliers to improve the design.
Because the sequence of developments in time is known, it becomes clear which levers were important at what moment in time. So, the sequence of developments indicates which levers receive the most attention at a certain point in time. From table 2 it becomes clear that the supplier focus of this manufacturer has shifted over time from the right side of the matrix (i.e. diagnostic control systems) towards the left side (i.e. interactive control systems).
However, in most cases these shifts do not mean that a previous lever gets no attention at all once the focus is on another lever. In most cases the manufacturers in this research kept existing quality management systems in place but felt the need to put more emphasis on different kinds of quality management systems in order to achieve their quality strategy.
Table 2 also shows that (for the example presented) two levers of the Simons model (i.e. beliefs systems and boundary systems) receive no attention because there is no quality management system in place in the organization that controls the suppliers from the perspective of these two levers. The absence of certain levers in the management of the quality of the manufacturer’s suppliers is clearly demonstrated by the matrix in table 2, which allows managers to think about the consequences of missing levers. In the example in table 2 there is no boundary system. So, the manufacturer has imposed no clear boundaries for the quality level of its suppliers. Apparently top management thinks that such boundaries are not necessary because the ranking systems 12
