Design Decisions and Co-operative Development of ...10385/FULLTEXT01.pdf · design and the...

Design Decisions and Co-operative Development of

Manufacturing Systems

A Doctoral Thesis by

Pär Mårtensson

TRITA-IIP-06-06 ISSN 1650-1888 © Pär MårtenssonSchool of Industrial Engineering and ManagementDepartment of Production Engineering Royal Institute of Technology S-100 44 Stockholm, Sweden Stockholm 2006, Universitetsservice US AB

iii

PREFACE

Research is very interesting. The science of artefacts is of particular interest since their origin is not driven by random natural reactions but rather by human conscious decisions and actions.

In the search for the ultimate proof of the logical laws of development one hits many interesting examples of products, processes and equipment. The continu-ous change makes life interesting but creates brain distortion when looking for stable fundamental truths. Another reflection made is the large amount of area specific knowledge, both formal and tacit knowledge one hits when going into deep discussions regarding manufacturing.

The last years I have been reflecting on human perception of the world and the possible risk for built-in incompleteness of our possible sensors and the restric-tions of our behaviour that is imposed by the fact that we act and build experi-ence.

At your hands is a thesis that I would have liked to work on for another fifty or so years.

iv

ACKNOWLEDGEMENTS

Research is not a work being performed by one person. During this research I was involved in many research and industrial projects. I would like to take the opportunity to thank the following people for their various contributions:

First of all Dr Sven Hjelm at Scania has made the research possible. Especially his outstanding capability of driving and finding new research and develop-ment directions has been helpful and inspiring.

Professor Bengt Lindberg has been guiding the second half of my research period as well as the writing of this thesis, always putting questions for trigger-ing my thoughts. Professor Gunnar Sohlenius guided the first half of my re-search period and gave a broad and philosophical view of the subject. Professor Torsten Kjellberg and Dr Mauro Onori have been helpful as co-advisors.

Over the years I had a nice and interesting time working with colleagues at different universities and companies. Thank you Mr Ingvar Johnson, Dr Per Gullander, Dr Pär Klingstam, Dr Derrick Tate, Dr Mattias Johansson, Dr Volker Plapper, Mr Jonas Rosén, Dr Johan Nielsen, Dr Peter Almström, Mr Jonas Fagerström, Mr Bruno Bernard, Mr Robert Lundqvist, Dr Johan Nielsen, Ms Astrid von Euler-Chelpin, Mr Patrik Holmström, Mr Mikael Lundkvist, Mr Gordon Sjöqvist, Mr Lars Johansson, Mr Erik Sigfridson, Mr Mats Boström, Mr Karl-Erik Rundlöf, Mr Erik Höppö, Mrs Petra Ålund, Mr John Du, Dr Lars Holst, Mr Jan Olof Svebeús, Dr Dario Aganovic, Dr Henric Alsterman, Mr Hasse Bengtsson, Mr Mats Söderberg, Mr Rolf Eriksson, Mr Michael Ekvall, Mr Ulf Bjarre and Mr Johan Nyberg.

Mr Mattias Söderholm and Dr Per Petersson have made a good effort in push-ing me towards the publication of this thesis, knowing the feeling of being within the academic knowledge creation process and at the same time within industrial activities.

All colleagues within the production engineering and consulting group at Scania have been standing my questions and always been responding with pro-fessional and friendly attitude.

All colleagues at the department of production engineering at KTH have con-tributed to a challenging and interesting environment touching various topics.

Thanks to Mr Andreas Rosengren for discussions and valuable illustrations.

This research was financed by Scania and Swedish Foundation for Strategic Research through the PROPER program.

v

ABSTRACT

This thesis presents contributions within two domains of manufacturing system development. The first domain is systematic design of product and manufactur-ing system. The second domain is model based co-operative development be-tween manufacturing companies and manufacturing system suppliers.

The axiomatic design framework for design is used to describe the product design and the manufacturing system design. The most important contributions are that product properties and analysis of product function are considered when specifying the manufacturing system. The systematic adaptation of prod-uct to the selected manufacturing solution is another contribution. A new con-cept of manufacturing phase functional requirements has been defined in order to cope with requirements from the manufacturing process on the in process product.

The co-operative concept is based on the use of ISO 10303-214 for specifica-tion and communication of proposed systems and equipment. The concept has been tested using real project information in prototype implementations.

Keywords: manufacturing system, design methodology, co-operation

vi

TABLE OF CONTENTS Abstract Table of contents List of figures viiiAbbreviations Term definitions List of publications

v vi

xii xiv xv

1 Introduction...............................................................................................2

1.1 Manufacturing of products .................................................................2 1.2 Engineering activities in product realisation......................................3

1.2.1 Product design.............................................................................4 1.2.2 Process planning .........................................................................4 1.2.3 Manufacturing system design .....................................................4

1.3 Specialisation and global division of work.........................................4 1.4 Globalisation and competition............................................................5 1.5 Facts on manufacturing industry in Sweden.......................................6 1.6 Problem definition ..............................................................................7 1.7 Limitations ..........................................................................................7 1.8 Hypothesis...........................................................................................7

2 Scientific Approach...................................................................................8

2.1 Scientific ideal.....................................................................................8 2.2 Research method.................................................................................9

3 Design and Manufacturing research.....................................................12

3.1 Manufacturing paradigms ................................................................12 3.2 Mass customisation and personalisation ..........................................14 3.3 Product architectures........................................................................15

vii

3.4 Product platforms .............................................................................16 3.5 Manufacturing system design ...........................................................16 3.6 Design models...................................................................................19 3.7 Development tools.............................................................................21 3.8 Term definition and information modelling......................................23 3.9 Analysis of state-of-the-art and motivation.......................................24

4 Design Decisions and manufacturing system design............................26

4.1 Product design decisions ..................................................................26 4.1.1 Product system structure definition...........................................26

4.2 Process planning...............................................................................33 4.2.1 Definition of in process view....................................................33 4.2.2 Process selection .......................................................................34 4.2.3 Process operational sequence and grouping..............................39 4.2.4 Process implications on product design ....................................41

4.3 Manufacturing system design ...........................................................44 4.3.1 System development scenario...................................................44 4.3.2 Manufacturing system specification .........................................46 4.3.3 Manufacturing system physical domain ...................................54 4.3.4 Information from supplier.........................................................55

5 Discussion and conclusions.....................................................................63

5.1 Discussion.........................................................................................63 5.1.1 Discussion on hypothesis..........................................................63

5.2 Scientific contributions .....................................................................66 5.3 Conclusions.......................................................................................67 5.4 Future research.................................................................................67

viii

LIST OF FIGURES

Figure 1. The role of manufacturing activities...................................................2 Figure 2. Manufacturing function [Sohlenius 2004]..........................................3 Figure 3. Engineering activities, selected feed back loops shown.....................3 Figure 4. Exports increase compared to production [Ekdal 2003].....................5 Figure 5. Composition of industrial production in Sweden year 2004,

share of total value added [Andersson-Stråberg A. et al, 2005]. .......6 Figure 6. Investments in Swedish manufacturing industries 1993-2005,

SEK billion, 2000 prices [SCB 2005] ................................................6 Figure 7. Induction and deduction [Chalmers 1995] ........................................8 Figure 8. Overview of the research work.........................................................10 Figure 9. Evolution of manufacturing paradigms [Jovane et al 2003].............12 Figure 10. Five principles to be used in order controlled customized

production [Sohlenius and Sandell 1975].........................................14 Figure 11. Four product architectures illustrated by trailer example

[Ulrich 1995]. ...................................................................................15 Figure 12. Variety as function of product architecture and process

flexibility. [Ulrich 1995]. .................................................................15 Figure 13. Manufacture of a modular built product principle [Erixon et al

1996].................................................................................................16 Figure 14. Reconfigurable line and 5-axis machining center [Koren

1999].................................................................................................17 Figure 15. Hyper flexible assembly system [Alsterman 2001]........................17 Figure 16. Costs analysis of a machining center [DESINA 2001]. .................17 Figure 17. Schematic and actual laboratory set-up of an evolvable

assembly system [EUPASS 2006]....................................................18 Figure 18. Design phase and design object based product design model

including process planning [Bolmsjö and Gustafsson in Randell 2002].................................................................................................19

ix

Figure 19. Design object based product design model including process planning [Suh 1995]. ........................................................................20

Figure 20. Systemic map for concurrent product and business process design [Sohlenius 2004]. ..................................................................20

Figure 21. Current product design development tools [Krause et al 2003]. ....21 Figure 22. Feature based concept for assembly planning [Bley and Franke

2004].................................................................................................22 Figure 23. Electrical and mechanical data integration [Rech 2004]. ...............23 Figure 24. Product system structure for product variety offered to

customers..........................................................................................27 Figure 25. Example of general case 1, standard product .................................28 Figure 26. Example of general case 2, segmented product..............................29 Figure 27. Crank mechanism. ..........................................................................30 Figure 28. Simplified logic for connecting rod................................................31 Figure 29. Three of the selected connecting rod design parameters. ...............31 Figure 30. Example of general case 3, parameterised product. .......................32 Figure 31. In process view with in process functional requirements and in

process design parameters. ...............................................................33 Figure 32. Assembly process dictated by product design solution. .................34 Figure 33. Friction affected by the manufacturing process, after [Xiao

2002].................................................................................................35 Figure 34. General surface conditioning process.............................................35 Figure 35. General geometry creating process.................................................36 Figure 36. Example of geometry creating process...........................................36 Figure 37. Example of material property process. Induction hardened

surfaces in red...................................................................................36 Figure 38. Selected process requires certain starting geometry. Starting

geometry requires a certain process..................................................37 Figure 39. Fine milling operation requires input geometry with small

variation. Acceptable variation can be achieved by rough milling. .............................................................................................37

x

Figure 40. General side effect of selected process causing recovery process. .............................................................................................38

Figure 41. Example of side effect of selected process. ....................................38 Figure 42. Example of side effect on workpiece condition. ............................38 Figure 43. Welding before drilling. .................................................................39 Figure 44. Drilling before welding. .................................................................40 Figure 45. In process requirements on in process product...............................41 Figure 46. In process design parameter violating product function.................42 Figure 47. Process constraints on product design. ...........................................42 Figure 48. Process simplification constraint on product design. .....................43 Figure 49. Same bearing diameter on camshafts for six and eight cylinder

engines..............................................................................................43 Figure 50. Communication links in manufacturing system development.

External product engineering competence necessary for the design of the manufacturing system in dashed. ................................44

Figure 51. General overview of a manufacturing system development process. .............................................................................................45

Figure 52. General process requirements on a manufacturing system.............46 Figure 53 Example of grouping work elements into operations. .....................47 Figure 54. Aggregated requirements on manufacturing system. .....................48 Figure 55. Required function on a spot welding cell by input/output

description. .......................................................................................49 Figure 56. Three independent cell functions....................................................49 Figure 57. Product information from manufacturing company. ......................50 Figure 58. Two alternatives of set ups of two different connecting rods.........51 Figure 59. Manufacturing system constraints affecting manufacturing

system physical domain....................................................................52 Figure 60. Manufacturing system design parameter. .......................................54 Figure 61. Example of time plan from supplier. ..............................................55 Figure 62. Two process descriptions including process and time...................56

xi

Figure 63. Example of functional description of a fixture. .............................57 Figure 64. Example of layout drawing including bill of material....................58 Figure 65. Design proposal from supplier. ......................................................59 Figure 66. Final design of in-station/ out-station.............................................59 Figure 67. Example of models enabling geometry simulation. .......................60 Figure 68. Archived information on one machine tool. ...................................62 Figure 69. Example of maintenance points on machine layout and

corresponding maintenance instruction. ...........................................62

xii

ABBREVIATIONS

AP – Application Protocol

ARM – Application Reference Model

BOM – Bill Of Material

CAD – Computer Aided Design

CAE – Computer Aided Engineering

DFM – Design For Manufacturing

DP – Design Parameter

EAS – Evolvable Assembly Systems

ENGDAT - ENGineering DATa Message

FR – Functional Requirement

FOD – Function Oriented Design

GDP – Gross Domestic Product

GNP – Gross National Product

IDEF0 – Integration DEFinition language 0

IDEF3 – Integration DEFinition language 3

IP – In Process

ISO – International Organization for Standardisation

M-DP – Manufacturing Design Parameter

M-FR – Manufacturing Functional Requirement

MIT – Massachusetts Institute of Technology

NC – Numerical Control

OSTN – Object State Transition Network

PDM – Product Data Management

PDTnet – Product Data Technology network

xiii

PLC – Programmable Logic Controller

PV – Process Variable

XML – eXtensible Mark-up Language

XSLT – eXtensible Stylesheet Language Transformations

STEP – STandard for the Exchange of Product model data

UOB – Units Of Behaviour

xiv

TERM DEFINITIONS

Capacity: The maximum processing, manufacturing or output per time unit [Peterson 2000].

Design: Industrial activity preliminary to the manufacturing of a product con-sisting of the conception and planning of the overall physical and performance characteristics of the product. [CIRP 2004]

Manufacturing system: Is an organisation in the manufacturing industry for the creation of production...Note...in English usage, the term “manufacturing system” addresses a complete enterprise or a group of enterprises, an individual production department or even a single work station. [CIRP 2004]

Operation: A defined action, i.e., the act of obtaining a result from one or more work elements (see also work element). [CIRP 2004]

Process planning: Includes all planning measures to be taken once to ensure the production of a part, an assembly or a final product at the lowest cost and best quality.[CIRP 2004]. In this thesis a more narrow definition is used where manufacturing system design is treated separately.

Product: Item resulting from an activity, operation or process, particularly from a manufacturing process. [CIRP 2004]

Production: The pure act or process of actually physically making a product from its material constituents. [CIRP 2004]

Work element: In process planning, it is a single task to be performed that cannot be subdivided. [CIRP 2004]

xv

LIST OF PUBLICATIONS

Research results in this thesis have partly been published in earlier publica-tions. Unabridged versions of paper A – D are appended. They contain infor-mation on referenced case studies and give the reader background information as well as a view of the evolution of this research work.

Paper A: Manufacturing subsystem design decisions,

Proceedings of the 33rd CIRP International Seminar on Manu-facturing Systems, 5-7 June 2000, Stockholm, Sweden.

Published in CIRP Journal of Manufacturing Systems

Volume 31, 2001, No. 1

Paper B: Product Function Independent Features in Axiomatic Design,

Co-authored with Jonas Fagerström

Proceedings of the First International Conference on Axiomatic Design, ICAD 2000, June 21 – 23 June 2000, MA, USA

Paper C: Functional coupling in manufacturing systems and its implica-tions,

Co-authored with Peter Almström

Proceedings of International CIRP Design Seminar – Design in the New Economy, 6 – 8 June 2001, Stockholm, Sweden

Published (compact version), A categorization of functional cou-plings in manufacturing systems, in Journal of Engineering Manufacture, Volume 216, No 4, 2002

Paper D: Co-operative digital projecting of manufacturing equipment – method and implementation based on STEP and XML,

Co-authored with Jonas Fagerström and Johan Nielsen

Proceedings of the 33rd International Symposium on Robotics, ISR 2002, October 7 – 11 2002, Stockholm, Sweden

xvi

Paper E: Communication for development of machining systems - method and implementation based on STEP and XML,

Not published

Paper F: Minimising the impact of changes and disturbances on manufac-turing system performance,

Co-authored with Jo Wyns, Patrick Peeters, Jörn Neuhaus and Annika Johansson

Proceedings of the second International Workshop on Intelligent Manufacturing Systems, Leuven, Belgium, 1999

Published in International Journal on Studies in Information and Control, Vol 9, No 2, 2000

Paper G: Reuse of Design Rationale in Cellular Manufacturing Systems: Theory and Application,

Co-authored with Derrick Tate

Proceedings of the 2nd Conference on Engineering Design and Automation, ED&A ´98, HI, USA, 1998

Paper H: Generic Modelling of Cell Control Systems and Production Re-sources,

Co-authored with Per Gullander and Pär Klingstam

Proceedings of the 7th Conference on Flexible Automation and Intelligent Manufacturing, FAIM ´97, 1997

Paper I: Reliability and Maintainability Consequences of Couplings in Manufacturing Systems,

Co-authored with Peter Almström and Torbjörn Ylipää

Proceedings of 35th CIRP International Seminar on Manufactur-ing Systems, Seoul, Korea, 2002

Licentiate Thesis: Conceptual Design of Manufacturing Subsystems,

Department of Manufacturing Systems, Royal Institute of Tech-nology, Stockholm, 2000

1

PART I Problem description and research approach

2

1 INTRODUCTION

The purpose of this introductory chapter is to describe the background and objective of the presented research. It defines manufacturing and shows the importance of manufacturing activities. One hypothesis is presented.

1.1 Manufacturing of products Manufacturing is an activity that is caused by the need of physical products. Products in turn, are things that allow the human beings to enjoy life by giving the user required and desired functions for usage. Key competitive factors are the ability to provide products with desired functions to low cost at the time there is a need for the function. Manufacturing has to be seen as part of a prod-uct lifecycle and must be performed within the constraints for sustainable life conditions on earth.

Figure 1. The role of manufacturing activities

Manufacturing originates from the Latin ”manus factus” which means ”the hand is making” [Sohlenius 1980]. This shows that the original meaning of the word was the process of performing something manually in the material world.

Manufacturing Recycling

Usage/ function provideri.e. transport function

3

Manufacturing can be defined as a transformation of material into products by changing the properties of the material by the addition of energy controlled by information. The addition of energy is carried out by exe-cution of processes using manufacturing resources. The manufacturing resources require capital and the de-velopment process as well as the manufacturing requires knowledge and skill. The information that controls the manufacturing process is refined information contain-ing market needs and tech-nical possibilities.

Figure 2. Manufacturing function [Sohlenius 2004]

1.2 Engineering activities in product realisation The manufacturing of products requires that there exists a manufacturing sys-tem. The manufacturing system has to be designed for manufacturing products by execution of defined processes. Product design is more or less constrained by existing or possible processes.

Figure 3. Engineering activities, selected feed back loops shown.

A0

Manufacturing

Material

Product

Working conditions

Waste backto nature

Society

Share holders

Market, customer needs

CapitalKnowledge and skill Energy

A1

Design product

A2

Plan process

A3

Designmanufacturing

system

Product models

Process plans, in-process models

Manufacturing systemmodel

Engineeringknowledge CAE systems

Customer needs

Needed volumes

Engineeringknowledge


CAE systems

CAE systems

Process constraints,existing or possible

processes

Equipmentconstraints,

existing or possible equipment

Field qualityNorms and regulations

A1

Design product

A2

Plan process

A3

Designmanufacturing

system

Product models

Process plans, in-process models

Manufacturing systemmodel

Engineeringknowledge CAE systems

Customer needs

Needed volumes



CAE systems

CAE systems

Process constraints,existing or possible

processes

Equipmentconstraints,

existing or possible equipment

Field qualityNorms and regulations

4

1.2.1 Product design The design of the product determines what functions the products are intended to have and what physical characteristics the product must have in order to deliver the intended functions. Design information is produced using knowl-edge from potential customers’ needs and technical possibilities provided by nature. Especially manufacturing process and equipment consequences are important to consider.

1.2.2 Process planning The process plan determines what transformation processes have to be per-formed on the material in order to create the physical characteristics stated by product design. In parts manufacturing process planning the necessary incom-ing material must be defined. In assembly process planning the assembly se-quence implies the incoming material. Furthermore the process parameters must be determined.

1.2.3 Manufacturing system design The design of the manufacturing system determines what functions the system is intended to have and what physical characteristics the system must have in order to deliver the intended functions. Manufacturing system design informa-tion is produced using input from product design, process planning and knowl-edge on technical possibilities.

1.3 Specialisation and global division of work Over the last decade, there has been a strong trend towards specialisation within automotive industry. The automotive companies focus on the engineer-ing design and manufacturing of the “vehicle” and the manufacturing system suppliers focus on the engineering design and installation of the “manufactur-ing system equipment”. Many companies seem to support the usage phase of their delivery (both vehicles and manufacturing system). The degree of system deliveries from tier one and tier two suppliers is also increasing. There are cases where suppliers offer the complete final assembly operations to different vehicle manufacturers [PricewaterhouseCoopers 2002]. These factors stress the importance of sharing engineering information between different organisational entities. One solution to the engineering communication problem is standard-ised communication of engineering information.

5

1.4 Globalisation and competition Global trade, global innovation and global production are now possible by, among others, the increase of international trade agreements, international standardisation and information technology. This implies sharper competition that forces companies to respond to customer needs quicker than before and still maintaining low cost and short delivery times. The total world trade has increased about three times the increase of world production over the last 50 years, see figure 4. This indicates that companies have to produce products meeting customer requirements from different markets. It also indicates that companies meet new competitors on their own home markets.

The needed ability to meet a wider range of requirements is stressing the im-portance of appropriate manufacturing processes and efficient production sys-tems.

Figure 4. Exports increase compared to production [Ekdal 2003].

W orldtrade and world production (GNP) in volume(1950=100)

1950 2000

100

500

1000

1500

2000

Export GNP

6

Engineering industry

54%

Other manufacturing

industry8%

Chemicals 15%

Food industry6%

Wood products and pulp, paper and graphics

17%

1.5 Facts on manufacturing industry in Sweden Industrial production comprised about 20% of GDP in 2004. GDP in year 2004 has been estimated to 2546x109 SEK [Ministry of finance, 2005]. The engineering industry creates about half of the total value added within indus-trial production sector, see figure 5 below. Also in terms of exports, the engi-neering industry comprises about 50%. The engineering industry is therefore important from a national perspective.

Figure 5. Composition of industrial production in Sweden year 2004, share of total value added [Andersson-Stråberg A. et al, 2005].

The investments in machinery corresponds to about 15-20% of the total value added and investments in buildings are about 3% of the total value added, see figure 6.

Figure 6. Investments in Swedish manufacturing industries 1993-2005, SEK billion, 2000 prices [SCB 2005]

Total

Machinery

Buildings

7

1.6 Problem definition The manufacturing system design sets the preconditions for the manufacturing activities. Investments in machinery for manufacturing are necessary and at the same time one major cost driver. It is therefore important for the competitive-ness and in a bigger context for social welfare, that investments are directed to value adding activities and maximise the value of manufactured products.

There is a need for methods for specification of appropriate manufacturing systems where products satisfying the customers’ needs can be achieved to low cost and short delivery time. For the realisation of the manufacturing system there is a need for co-operation with suppliers. These two needs can be formu-lated into one over-all research question that has been guiding the research.

Research question: What method could be used for improved development of manufacturing systems?

1.7 Limitations The scope for the thesis has been limited to processes for manufacturing of advanced mechanical products at medium high volumes by available case study objects. The case studies have mainly been performed within automotive indus-try within heavy truck assembly and machining. The researcher has had a pro-duction engineering view with technical focus. Economical and social aspects are not detailed.

1.8 Hypothesis Considering the importance of appropriate manufacturing systems and the globalisation of engineering activities involving computer supported tools one hypothesis for successful manufacturing system design has been formulated:

Hypothesis:

The understanding, capturing and structuring of (in process) process fea-tures may lead to a fruitful connection to product design aspects such that more effective manufacturing system development is achieved.

8

2 SCIENTIFIC APPROACH

This chapter describes the author’s view of science and how the research has been performed.

2.1 Scientific ideal Scientific knowledge is proven knowledge. By using scientific knowledge one can make predictions for the specific case. Science is objective. This view of science is called inductionism [Chalmers 1995] and is based on the assumption that there exists cause and effect.

Figure 7. Induction and deduction [Chalmers 1995]

The scientific ideal above is difficult to reach in all research domains and could be seen as a vision for this research work. One problem, originally formulated by Kaplan in 1964, is the paradox of sampling. The paradox states that a sam-ple has to be truly representative for the whole population. However the sample is only representative when we know the characteristics of the whole popula-tion in which case we have no need of a sample. [Handfield & Melnyk, 1998] This means that the validity of the theory cannot be known. The probability of truth can be increased by increasing the number of observations and samples. Karl Popper reversed the reasoning into:

• If the theory is true, then the prediction is true.

• The prediction is not true.

• Therefore, the theory is not true.

Induction Deduction

Theory

Observation Prediction, explanation

9

Popper had the view that there exists a truth and that science develops all the time towards the truth.

The engineering discipline is more problem oriented than the natural science. Engineering is not only to understand and explain existing natural phenomena. It is also involving the creation of new things. H. von Karmann at MIT once expressed this in the following way: “The scientist explores what is and the engineer creates what has never been.” Here science means natural science.

Sohlenius states: “The knowledge about what is possible we obtain from stud-ies within the natural and technical sciences. Within applied subjects the knowledge about what is possible must be combined with knowledge about what is desirable. What is desirable is a question belonging to the human sci-ences. Engineering science therefore is a bridge between natural and technical science on one hand and human science on the other.”[Sohlenius 2002]

2.2 Research method Sohlenius has developed a scientific method for research and development in science of engineering, where science of manufacturing engineering is one part. The method consists of six steps:

• Analyze what is

• Analyze the possible

• Define the desirable

• Develop what has never been

• Analyze the result

• Generalize, extend theories

This research has been performed with these steps in mind. The case study strategy has been used. It is not possible to state the generality of the hypothe-sis and method details due to the small number of case studies. In figure 8 the performed case studies are shown. The case studies had three purposes.

- Observations of real manufacturing systems and engineering projects for understanding the problem area as well as for identification of theo-retical gaps and formulation of hypothesis.

- Verifications of hypothesis.

- Validations of hypothesis in new cases for determination of generality.

10

1

Explorative casestudy of rear axle

project

2

Explorative casestudy pinionmachining

3

Development ofhypothesis

5

Validation caseconnecting rod

project

Research question onmethods for developmentof manufacturing systems

6

Explorative casestudy cabwelding

”...understanding, capturing andstructuring of (in-process)

process features…"

7

Development ofhypothesis

”...may lead to a fruitful connection to product design aspects…"

9

Validation caseengine block line

Needs for communication betweenmanufacturing company and supplier

Need for decision basisfor specification of equipment

Importance of processrequirements on product

8

Verification, cabwelding

4

Verification, rearaxle and pinion

Verifyed part of hypothesis

Validity of part of hypothesis

Verifyed part of hypothesis

Validity of part of hypothesis

Figure 8. Overview of the research work1.

1 Research work is here considered as the work performed within the planned case studies. The author has been active in two engineering projects regarding gear box shaft straightening and cam shaft machining that have given additional knowledge and information. The projects where not part of the planned research and there are no published academic results.

11

PART II

Frame of reference

12

3 DESIGN AND MANUFACTURING RESEARCH

This chapter presents selected parts of background information and existing theory that is important for the results in section III.

3.1 Manufacturing paradigms The way companies have been acting has changed over time, driven by the market and societal needs along with technology development. Four consumer goods manufacturing paradigms can be identified in modern times and a fifth coming paradigm is emerging. [Jovane et al 2003]

Paradigm Craft production

Mass Production

Flexible Production

Mass Customization

and Personalisation

Sustainable Production

Paradigm started

~ 1850 1913 ~ 1980 2000 2020?

Society Needs

Customized products

Low cost products

Variety of products

Customized products

Clean products

Market Very small volume per

product

Demand > Supply

Steady demand

Supply > Demand

Smaller volume per product

Globalization

Fluctuating demand

Environment

Business Model

Pull sell-design-

make-assemble

Push design-make-assemble-sell

Push-Pull design-make-sell-assemble

Pull design-sell-

make-assemble

Pull design for envi-ronment-sell-

make-assemble

Technology Enabler

Electricity Interchangeable parts

Computers Information Technology

Nano/Bio/Material Technology

Process Enabler

Machine Tools Moving Assembly line &

DML

FMS Robots RMS Increasing Manufacturing

Figure 9. Evolution of manufacturing paradigms [Jovane et al 2003].

The craft production paradigm was based on manufacturing one-off products according to individual specifications starting with a sell-process. Flexible ma-chines and highly skilled workers made it possible to design and manufacture to order.

13

The mass production paradigm was based on manufacturing high volumes of identical products where the low price made the market unlimited.

The flexible production paradigm was based on requests of more diversified products. Parts were produced according to mass production principles but order based variant creation in assembly.

Mass customisation and personalisation is the current theme in consumer product manufacturing industry and in some other industries. Supply exceeds demand and one way to compete is to offer the product that each customer wants. Information technology makes it possible to have a dialog with each individual customer.

Sustainable production is an emerging paradigm that is driven by the societal need for a better environment. The whole value chain as well as the usage phase and disposal of the product are being consciously considered. The re-use of production equipment is one important factor.

14

3.2 Mass customisation and personalisation There are five principle ways to handle the order phase for manufacturing of customised products. The principle ways differ in the degree of preparation and the order dependent work to be carried out [Sohlenius and San-dell, 1975], see figure 10.

The structure can be seen as a basis for the research work de-scribed in this thesis in the sense that the proposed method details the fourth principle way, “De-sign to given rules, parameter-ised product model” in combina-tion with “Process planning as to rules”. Furthermore there is a need to connect the design deci-sion logic in the lower part of the figure to the product design decisions and process planning decisions.

Figure 10. Five principles to be used in order controlled customized production [Sohlenius and Sandell 1975].

The last years increased use of internet technologies has given new possibilities to develop the customer interface permitting the customer to participate in the design process [Tseng et al 1996], [Jovane et al 2003]. From designing prod-ucts companies move towards designing the systems that provide the products and services [Tseng et al 2003].

Customer

Bid Order

Order specification

Product break downMain design

Storedcomponents

Componentspurchased toeach order

Drawings orproductmodel

available

Design to givenrules

Parameterisedproduct model

New design

Processplans

available

Processplanning

as torules

Processplanning

as torules

Processplanning

Manufacturing documents

Parts manufacturing to order

Manufactureor

purchase onforecast

Manufactureproduct type

Part familymanufacture

Functionalworkshop

Assembly planning

Assembly

Unique product

15

3.3 Product architectures The scheme by which mapping of functions into physical components is made is sometimes referred to as product architecture. Ulrich defines four principle architectures where three are modular and one is integral. The three modular schemes differ in the degree of freedom in the interfaces between modules.

Figure 11. Four prod-uct architectures illus-trated by trailer exam-ple [Ulrich 1995].

Ulrich states that much of the competitiveness of the company depends on the selection of product architecture. He also states that integral designs might be as suited as modular architec-tures for variety creation for cus-tomised products.

This depends on the process char-acteristics in component manu-facturing and assembly. [Ulrich 1995]

Figure 12. Variety as function of product architecture and process flexibility. [Ulrich 1995].

Mod

ular

• Variety achieved by combinatorial assembly from relatively few component types.

• Can assemble to order from component inventories.

• Minimum order lead time dictated by final assembly process.

• May fabricate components to order as well as assemble to order.

• May choose to carry component inventories to minimize order lead time.

• Infinite variety is possible when components are fabricated to order.

• High variety not economically feasible; would would require high fixed costs (e.g. tooling), high set-up costs, large order lead times, and/or high inventory costs.

• Variety can be achieved without relatively high inventory costs by fabricating components to order.

• Minimum order lead times dictated by both component fabrication time and final assembly time.

• Infinite variety is possible.

Low High

Prod

uct A

rchi

tect

ure

Inte

gral

Component Process Flexibility

16

3.4 Product platforms Product platform management addresses innovation over time and targets the product development including components, knowledge and production proc-esses. The trade-off between distinctiveness and commonality is the major is-sue. [Robertson et al 1998]

Similar thoughts are also described by Erixon showing “products in the prod-ucts – factories in the factory” and by Sjöström in the extensive paper on truck manufacturing. [Erixon et al 1996], [Sjöström 1991]

Figure 13. Manufacture of a modular built product principle [Erixon et al 1996].

3.5 Manufacturing system design The design of manufacturing system is a function of

- Requirements and constraints on the system

- Feasible technology

Dedicated systems and flexible systems fulfil specific requirements where the later has a wider possible product range. For meeting unknown requirements and for adopting new technologies the concept of reconfigurable systems has been developed. [Koren et al 1999], [Alsterman 2001].

17

Figure 14. Reconfigurable line and 5-axis machining center [Koren 1999].

Figure 15. Hyper flexible assembly system [Alsterman 2001]

One enabler for reconfigurability is standardisation of electric and fluidic inter-faces across vendors. The German machine tool industry has developed a Dis-tributEd Standardised INstAllation technology, DESINA. The costs for electric system and fluidic system are higher than for the mechanics in modern machine tools and are suited for standardisation. [DESINA 2001]

Figure 16. Costs analysis of a machining center [DESINA 2001].

Material and production Design and 76% development

24%Total costs

Electrical Fluidic Mechanicalequip. equip. equip.

34,2 % 7,6 % 34,2%

Material andproduction costs

18

In order to cope with changing requirements reconfigurability is a necessary but not sufficient system property. Ongoing research is now developing system architectures for the management of change in a wider sense than the physical system. Evolvable assembly systems, EAS, are built on a process-oriented ap-proach in order to enable a separation of product design and the equipment. When product changes or product shifts goes outside the range of the defined process, or the equipment skills, the equipment has to be re-engineered. The EAS concept includes planning of change and requires built-in change man-agement.

No real demonstrators have yet been developed, however feasibility of key enabling technologies is currently being evaluated in laboratory set-ups. [Onori et al 2006] , see figure 17.

Figure 17. Schematic and actual laboratory set-up of an evolvable assembly system [EUPASS 2006].

19

3.6 Design models Design models can be divided into at least three types:

- description of design phases

- description of engineering activities

- description of the design object

The models can be general or domain/ task specific. There exists a large amount of design models. 19 design models are described by [Evbuomwan 1996]. Most models describe the design of products where the process planning and manufacturing system design are not considered explicitly. The interface between product design and process planning is described in some models, see figures below.

Figure 18. Design phase and design object based product design model including process planning [Bolmsjö and Gustafsson in Randell 2002].

20

Suh [Suh 1990] models the background information concerning the design object from customer attributes and functional requirements and relates deci-sions in the physical domain to the process domain. Feed back information between domains for further decompositions is proposed, see figure 19.

Figure 19. Design object based product design model including process planning [Suh 1995].

Sohlenius [Sohlenius 2004] combines engineering design and manufacturing system design in his systemic map. Suhs’ design theory is complemented with business requirements and manufacturing processes and installation.

Figure 20. Systemic map for concurrent product and business process design [Soh-lenius 2004].

21

3.7 Development tools The research and development on computer based development tools started in the 1960’s. Big companies spent thousands of man years in software develop-ment. In the 1970’s commercial CAD systems (turnkey drawing systems) be-came available [Wozny 1989]. New development tools have been developed and most of them cover the physical aspects of products, not conceptual phase.

Figure 21. Current product design development tools [Krause et al 2003].

New development tools have been developed to support the early stages of product development. One example is the function oriented design system, (FOD), supporting designers when:

- setting up requirements

- specifying functions and function structures

- defining the part structure and its components

- managing nets of parameters and constraints

FOD is proposed to be used for process planning for the design object “produc-tion plant”, however this is a future research task. [Krause et al 2003]

22

Preconditions for industrial usage of advanced analysis tools have changed after the mergers within the CAD/CAE/CAM-sector that is leading to highly integrated solutions. Modules for support of design, process planning, manu-facturing system design and measurement in production can work on the same data, even on a feature level where the commercial STEP-exporters are not capable to represent the information, see figure 22 below.

Figure 22. Feature based concept for assembly planning [Bley and Franke 2004]

23

3.8 Term definition and information modelling ISO 10303 is an international standard for the computer-interpretable represen-tation and exchange of product data. The objective is to provide a neutral mechanism capable of describing product data throughout the life cycle of a product, independent from any particular system. The AP 214 (application protocol) is addressing the information exchange requirements when develop-ing vehicles (cars, trucks, buses, and motorcycles) [ISO 2001]. The AP 2402 is addressing the information exchange requirements when planning manufactur-ing where parts are machined [ISO 2004].

It has been shown that the ISO 10303-214 covers the basic information ex-change when developing manufacturing systems [Johansson 2001]. Usefulness and additions for covering the needs for process planning has been proposed by [Nielsen and Kjellberg 2000] and [Nielsen 2003]. The approach to utilise STEP standards for the manufacturing systems development has been further devel-oped within the IDA-STEP project. [Rech and Klein 2001], [Rech et al 2004].

Figure 23. Electrical and mechanical data integration [Rech 2004].

2 Published as standard 2005-12-09 during fine tuning of this thesis.

24

3.9 Analysis of state-of-the-art and motivation Research within manufacturing processes and physical manufacturing system develops new technologies. Usually there is no or little design methodology described in relation to the new processes and technologies. For a manufactur-ing company it is of outmost importance to achieve both quality and productiv-ity over time. To achieve this it is necessary with an overall picture as well as elaborated details. Depending on pre-conditions in each case, certain processes and certain technologies are more or less suitable. Therefore work methods for development are important. These methods can be described by design models, such as in section 3.6.

Few design models cover product design, process planning and manufacturing system design. Some found exceptions are described briefly in section 3.6.

The models by Bolmsjö and Gustafsson, and also by Suh, have no manufactur-ing system design aspect but product design and process planning aspects only.

The model by Sohlenius has product design, process planning and manufactur-ing system design aspects. The model is abstract and illustrates on a general level the relations between elements within the different domains and between the domains themselves. It also relates some existing theories and work meth-ods to the domains. In order to understand and make use of the model one has to know the underlying theories and methods.

There is a gap between the model and how to proceed when developing manu-facturing systems. Therefore this thesis is going deeper into the definition of some necessary elements and also illustrates these by examples found in differ-ent case studies. Some examples have been captured outside defined case stud-ies.

The fact that manufacturing system development is a work where the needs of the manufacturing company have to be understood, described and communi-cated to suppliers of machinery, there is a communication/co-operation activity to be performed. There has been a need to verify the usefulness of standardised terms and information models described in section 3.8.

Formal mathematical models of processes and the relation to product design aspects are still missing.

25

PART III Scientific contribution and result

26

4 DESIGN DECISIONS AND MANUFACTURING SYSTEM DESIGN

This chapter describes the resulting methodology for structuring the product offer in order to specify the manufacturing system. The statements are based on observations from the case studies, complemented with examples from litera-ture.

4.1 Product design decisions The product design has several implications on the manufacturing system de-sign and must be made in accordance with possible manufacturing processes.

4.1.1 Product system structure definition In the planning of products, functional requirements on the product are set. In order to set the functional requirements on the product one has to understand the underlying customer needs and not only the product operation but also the life conditions including serviceability and disposal. In addition to functional requirements based on customer needs, laws and regulations affect the product design by imposing new functional requirements and/ or constraints on the product design.

Statement 1: Customer needs can be grouped in three general cases:

a) All customers have the same needs3

b) Groups of customers have the same needs

c) All customers have different needs

3 One special case is where the customer need is reinforced by functional requirements laid down by law.

27

Statement 2: For each general customer needs case, a corresponding general product case can be defined:

a) One standard product

b) Segmented product, selection of variant or feature

c) Parameterised product

Statement 3: For each general product case, a corresponding general process case can be defined:

a) One standard process

b) Order dependent variant process

c) Order dependent parameterised process

Statements 1, 2 and 3 above builds up a system structure, see figure 24. The tolerance of the functional requirements can theoretically be different and thus making the system structure more complex.

Figure 24. Product system structure for product variety offered to customers.

FR 1 value x ± a

FR 1 value x ± a

FR 1 value y ± a

FR 1value x ± a

DP 1 PV 1

FR 1range x,y

DP 1parameter

PV 1parameter

General Case 3:Parameterised product

(order dependent process variable)

FR 1value x

DP 1 a PV 1 a

General Case 2:Segmented product

(selection of variant or feature)

General Case 1:One standard product

Customerusage type

FunctionalRequirement

FR 1value y

DP 1 b PV 1 b

Physicalproduct and

processrelation

FR 1 range x,y ± a

28

Observation supporting statement 1a, 2a and 3a:

Example of standard product/part

No standard products have been observed in any of the case studies. In order to find standard products/parts within the product range one has to go deep into assemblies. One example of standard parts is the screws for assembly of cylin-der heads. The screws are however bought from an external supplier and their manufacturing processes have not been studied, see figure 25.

Figure 25. Example of general case 1, standard product

29

Observation supporting statement 1b, 2b and 3b: Example of segmented product, selection of variant

For the module studied in the rear axle case [Mårtensson 2000] the important functional requirement to support the driven wheel axle, given a certain load, has been grouped in three classes4. For each class there is one main variant of axle. This leads to tailored process variables for each variant. In figure 26 there is a simplified description of high level process variables.

Support drivenwheel axle withaverage load

y ± a

Segmented product(selection of variant or feature)

Support drivenwheel axle with

high load z ± a

Average load on tractor

High load on truck

Support drivenwheel axle with

small load x ± a

FR 1 Support........ x ± a

DP 1 Axle withshaft endingdiameter 80

PV 1ABlanks A

Processes A

Small load on tractor

FR 1 Support........ y ± a


FR 1 Support........ z ± a


PV 1BBlanks B

Processes B

PV 1CBlanks C

Processes C

Figure 26. Example of general case 2, segmented product.

One important design parameter is the bearing solution. The different loads give different requirements on the bearing solution and the bearing solutions were here decided to be different. As a consequence the shaft endings are dif-ferent in terms of diameter and length. Furthermore the function to hold a hub reduction has been integrated into the 100 mm shaft ending and this compli-cates the picture.

4 Three classes and three variants at the time for the study, at the publishing time for this thesis the number is two.

30

Observation supporting statement 1b, 2b and 3b: Segmented product, part/ feature level

Within engines there is a need to transform the translational energy to rota-tional energy where one solution is the crank mechanism, see figure 27.

Figure 27. Crank mechanism.

In the connecting rod case study it could be observed that the requirements on the connecting rod depends upon the energy to be transformed. The design is subject to constraints on longevity and there is a need for minimisation of weight. The connecting rod as isolated item is not subject to major changes driven from customer needs as long as the engine concept is not changed. However, different engine variants require different connecting rods.

In figure 28 there is a simplified description of the system design logic includ-ing process domain.

Figure 28. Simplified logic for connecting rod.

The three design parameters in figure 29 are all de-pendent on the energy to be transformed from the piston to the crank shaft. The distance DP 1.2.1.3 is derived by a constrained ratio to the crank radius for achieving desired movement behaviour of the piston that in turn gives conditions for the combustion proc-ess. The crank radius can vary between engine vari-ants.

For further reasoning on the design parameter variation implication on the manufacturing equip-ment, see section 4.3.3.

Figure 29. Three of the selected connecting rod design parameters.

FR 1generateenergy

DP 1diesel engine

PV 1Engine

Manufacturingprocess

FR 1.2.1transformforce frompiston to

crank shaft

FR 1.3FR 1.1FR 1.2

transformpreassure to

rotation

Constraints fromregulations/ norms

DP 1.2crank

mechanism

DP 1.2.1connecting

rod

DP 1.2.1.1diametersmall end

(givesrequiredsurface)

DP 1.2.1.2diameterbig end

(permits therequired

dimension ofcrank shaft

DP 1.2.1.3distancebetween

holes (givesrequired

movementbehaviuor of

piston)

FR 1.2.1.1transfer

energy fromgudgeon pin

(surfacepressure)

FR 1.2.1.2transfer

energy tocrank shaft

FR 1.2.1.3transmitenergythrough

connectingrod

PV 1.2Crank

manufacturingprocess

PV 1.2.1Connecting rodmanufacturing

process

PV 1.2.1.1machinehole with

diameter DP1.2.1.1

PV 1.2.1.2machinehole with

diameter DP1.2.1.2

PV 1.2.1.3position ofsmall endrelative to

the big end

31

32

Observation supporting statement 1c, 2c and 3c: Parameterised product

The distance from the rear axle to the end of the frame is a parameterised de-sign parameter at the company where the rear axle case [Mårtensson 2000] was performed. The underlying customer need is to transport load and the load states requirements on the chassis. In order to adapt the transport solution to the customer need, specialised body-building companies design and manufacture the body. Depending on load distribution and manoeuvrability requirements a needed rear overhang can be determined. The length of the frame is then ad-justed by setting parameters for cutting the frame, see figure 30.

Enable space for load FR1 Enablespace for load

DP 1 Rearoverhang J(1 cm steps)

PV 1Length value in

cuttingmachine

Single customer,need for load space Parameterised product

Figure 30. Example of general case 3, parameterised product.

33

4.2 Process planning The process planning gives as an output the processes to be used for the manu-facturing of a single part or assembly and in what operational sequence the processes can be performed. The process planning is necessary for the specifi-cation of equipment used for the process as well as for specification of the in-coming material, parts or subassemblies. The decision on processes gives nec-essary input for the detailed design of products. It is important to understand the processes fitness for the required product volumes.

4.2.1 Definition of in process view It has been found that there is a need for an in process view of the product when reasoning on process planning and not only the final part or product [Mårtensson and Fagerström 2000], [Almström and Mårtensson 2001]. For this purpose an in process view has been introduced. The in process view captures explicitly the needs within the process as well as product adaptations that meet those needs. The in process view can work on top of original axiomatic design domains. In this section both in process functional requirements and in process design parameters are used, see figure 31 below. In process process variables are here treated as process variables. This notation is used where applicable and needed below.

FR1

IP FR1

DP1

Functionaldomain

IP DP1

PV1

PV1.1

In processfunctional

domainPhysical (final)product domain

In processproduct domain

Processdomain

Figure 31. In process view with in process functional requirements and in process design parameters.

34

4.2.2 Process selection Statement 4: Processes are either required for manufacturing the required de-sign parameters or required by other processes. Six groups of processes have been identified.

a) Assembly process

b) Functional surface process

c) Geometry process

d) Material property process

e) Conditional processes

f) Parts manufacturing recovery processes

Observation supporting statement 4a: Assembly process

For products that are more complex than consisting of one part an assembly process has to be selected. Typically the joint solution selection affects the product design and has to be made early in the product design process. First of all the joint solution has to fulfil the functional requirements on the product. Secondly the processes for performing the assembly have to be possible to perform. Once the design parameter for joining is determined the joint solution gives lower level functional requirements and constraints on parts to be joined and the selection of process gives constraints on the product and process. The assembly process selection is dictated by the product design decisions. In the cab welding cell case study [Mårtensson, Fagerström and Nielsen 2002] the spot weld joint was decided as design parameter for joining the parts and the spot welding process was chosen for the assembly of the parts, see figure 32.

Figure 32. Assembly process dictated by product design solution.

DP1 Part 1PV 3

Spot weldingprocess

FunctionalRequirement

Physicalproduct and

processlogic

DP2 Part 2FR1 Something FR2 Something DP3 Spot weldjoint

FR3 Join DP1and DP2

permanently

Process decision dictatedby product designparameter decision

35

Example supporting statement 4b: Functional surface process

Tribological properties can be controlled by creation of designed surfaces using certain processes. No studies on functional surfaces have been carried out within the case studies. The example in figure 33 shows friction as function of the Herzey parameter for two grinded surfaces and one phosphated surface. [Xiao 2002].

Figure 33. Fric-tion affected by the manufacturing process, after [Xiao 2002].

In general the surface character is a design parameter intentionally defined using R-parameters5 and certain processes have to be used for achieving the surface such as grinding, honing and polishing.

Figure 34. General surface conditioning process.

5 S-parameters are not ISO-standard at the publication date of this thesis

DP1Surface character A

PV1ProcessY

log H (m)

Fric

tion

coef

ficie

nt

-1-2

0,1

G-mediumG-fineF3 (Phosphated)

0

36

Observation supporting statement 4c: Geometry processes

The creation of specified geometries can be made by a wide range of processes. First of all required feature types exclude processes. Secondly both nominal values (dimensions) and geometrical tolerances determine the process candi-dates.

Figure 35. General geometry creating process.

In the pinion case [Mårtensson and Fagerström, 2000] there was a need to cre-ate geometry for splines. The cold rolling process was selected to create the specified splines geometry.

Figure 36. Example of geometry creating process.

Observation supporting statement 4d: Material property processes

Material properties depend on the material composition and microstructure as well as residual stresses created during the manufacturing process.

In the crank shafts case [Mårtensson and Almström 2001] induction hardening was selected for obtaining the specified material structure on certain features on the product.

Figure 37. Example of material property process. Induction hardened surfaces in red.

DP1Geometrical

Feature 1(nominal values,

tolerances)

PV1ProcessY

Nominal value(systematic error,

randomdistribution)

DP1Splines geometry

PV1Cold rolling

DP1Material

structure

PV1Inductionhardening

37

DP1GeometryFeature X

PV1Milling

(fine cut)

IP DPGeometryFeature Y

PV1.1Roughmilling


Physical (final)product domain

Processdomain

Observation supporting statement 4e: Conditional processes

A process might require starting conditions such as existing reference points, clamping surfaces or certain accuracy on the starting geometry. Furthermore a process might be inefficient for performing a process from possible blank or raw material. If the purchased material cannot fulfil these requirements there has to be additional processes performed. For the creation of these one must introduce conditional processes. These are not coupled to the product final design but rather to selected processes.

Figure 38. Selected process requires certain starting geometry. Starting geometry re-quires a certain proc-ess.

In the engine block machining case study [Mårtensson 2003], [Appendix C], the fine milling operation is preceded by a rough milling operation due to inac-curacy of the casting process and the required conditions for fine milling opera-tion.

Figure 39. Fine milling operation requires input geometry with small variation. Acceptable variation can be achieved by rough mill-ing.

DP1 PV1

IP DP PV1.1



Process domain

38

Observation supporting statement 4f: Parts manufacturing recovery proc-esses

Processes might have side effects and therefore require additional process to recover the damage or restore a condition, see figure 40.


PV1Process A

PV1.1Recoveryprocess B



Processdomain

Functionalrequirements


requirements

FR1Function 1

Non-wantedfeature orcondition

Figure 40. General side effect of selected process causing recovery process.

In the rear axle case [Mårtensson 2000] there was a need to perform deburring as a consequence after drilling and threading, see figure 41.

DP1Threaded Hole

PV1Drilling

PV 1.1Deburringprocess



Processdomain



requirements

FR1Enable joiningof differentialhousing and

rear axlehousing

Burrs

Figure 41. Example of side effect of selected process.

From the case study on crank shaft manufacturing there was an observed need to perform cooling operation as a consequence after milling [Almström & Mårtensson 2001], see figure 42

Figure 42. Example of side effect on workpiece condition.

PV1Milling

PV 1.1Coolingprocess


Processdomain

Heat insidework piece

39

4.2.3 Process operational sequence and grouping Going from a set of process variables, as work elements, to a defined sequence of operations involves understanding of the different processes and possibilities in equipment. In assembly also the product design itself imposes constraints on possible operational sequence.

Statement 5: Process operational sequence and grouping can be seen as the activity between process domain of the product and the functional requirement on the manufacturing system.

Observation supporting statement 5: Process operational sequence and grouping.

The transmission output shaft part consists of two parts that are assembled by welding. The base and the cover have holes that together position shafts for planet wheels. For minimisation of functional requirements on the equipment and maximisation of probability of success the welding is performed before the drilling, see figure 43.

Figure 43. Welding before drilling.

UoB/Assemblebase and cover UoB/Drill holes

State 1 State 2 State 3

40

An alternative sequence of operations would be to drill the holes before assem-bly, according to figure 44 below. The requirements on the assembly operation are here several magnitudes higher compared to the above case. There is also one extra operation that implies that there is one extra functional requirement on the equipment. These together give a lower prob-ability of success.

Figure 44. Drilling before welding.

The axiomatic design procedure for coupling analysis can be used, but in order to handle coupling between design parameters and in process design parame-ters the extension with in process functional requirements and in process design parameters has to be used.

Basestate 1

Basestate 2

Finalstate

UoB/Drill holes UoB/Assemblebase and cover

Coverstate 1

Coverstate 2

UoB/Drill holes

&

41

4.2.4 Process implications on product design Adapting a product design to the manufacturing process is known as DFx where x can be assembly, machining, forging etc. The process domain deci-sions have different implications on the product design.

Statement 6: The process domain decisions influence the product design in three ways:

a) by imposing in process requirements on the in process product

b) by natural constraints on the product design parameters

c) by cost reduction through process simplification

Observation supporting statement 6a: In process requirements on in proc-ess product

A process might require starting conditions of the in process product such as existing reference points, clamping surfaces or certain accuracy on the starting geometry. The requirements for such parameters are here called in process functional requirements and the parameters are called in process design pa-rameters.


PV1Process A

IP DPGeometryFeature Y

PV1.1Process B



Processdomain



requirements

IP FR1Clampable

part

FR1Function 1

Figure 45. In process requirements on in process product.

In many cases it is possible that in process design parameters are remaining on the final product without disturbing the product function or violating the prod-uct constraints. When existing design parameters are used as in process design parameters a coupling is created, see appended paper A [Mårtensson and Fagerström 2000]. In the cab welding case study it was observed that an in process design parameter is used in order to ease the requirements on fixtures and handling. In the middle of the windscreen frame reinforcement material is

42

left through the process, see figure 46. The reinforcement violates the product function and has to be taken away in later operations.

Figure 46. In process design parameter violating product function.

Observation supporting statement 6b: Process natural constraints on product design parameters

A process might constrain the product design. The constraints are related to possible features, possible dimensions or possible surface structure. The con-straints arise from the process selection and affect the physical product domain decisions, see figure 47.


FR1Function 1

DP1Part1

PV1Process A

Processdomain

Physical productdomain

DP1.1FeatureX

DP1.2FeatureY

Process naturalconstraint:

Possible featuresMinimum dimensionsMaximum dimensions

Variation of dimensionsPossible accuracy

Figure 47. Process constraints on product design.

Engine block design is constrained by natural constraints from the casting process. Variations in dimensions give conditions for generation of both resid-ual stress and pores due to cooling conditions imposed by the variations in dimensions. One example of minimal dimensions is that small radius on nega-tive corners gives unfavourable cooling conditions that might lead to porosity near the cast surface with a possible leakage effect. [Svensson 2005]

43

Observation supporting statement 6c: Simplification constraints on prod-uct design parameters

Process simplification is possible by reduction of number of different design parameter values where variance cost resources and/or effort. Feature types, dimensions and directions are examples of candidates for standardisation.

DP1Part1

PV1Process A

Processdomain

Physical productdomain

DP1.1FeatureX

DP1.2FeatureX

DP1.1.1Feature

dimensionY

Process simplificationconstraint:

Standard featuresStandard dimensionsStandard directions

DP2Part2

DP1.2.1Feature

dimensionY

DP2.1FeatureX

PV2Process A

DP2.1.1Feature

dimensionY

Figure 48. Process simplification constraint on product design.

It has been observed that all bearings on all crankshafts have the same diameter having, as an effect, that the process variables can be the same, see figure 49.

Figure 49. Same bearing diameter on camshafts for six and eight cylinder engines

7×

5×

44

4.3 Manufacturing system design Manufacturing system design gives as an output a description of a manufactur-ing system. Manufacturing system design is specified by the manufacturing company and various suppliers contribute with equipment design on different levels.

4.3.1 System development scenario A manufacturing system is the result of several co-operating organisational entities, see figure 50 below. The manufacturing company is the end user and specifies and operates the system. Equipment suppliers can be system suppliers and/or component suppliers. Some suppliers offer product design and process planning competence in order to support the manufacturing company in prod-uct development. In some cases it is necessary to involve the product develop-ment suppliers, both part suppliers and engineering resources.

Manufacturingcompany

Manufacturingsystem supplier

Manufacturingcomponent

supplier

Part supplier

Engineeringconsultant(product)

Figure 50. Communication links in manufacturing system development. External prod-uct engineering competence necessary for the design of the manufacturing system in dashed.

For rapid and precise development work it is necessary that the involved par-ties can reach information and publish their results so that work that is depend-ent on input from each other can go on without time loss or quality loss.

45

The information exchange requirements arise from the fact that the manufactur-ing company and the manufacturing system supplier have to co-operate in a common development process, see figure 51 below.

Figure 51. General overview of a manufacturing system development process.

Tenderrequest

Tender

Order

Detailed design andmanufacturing

Installation



46

4.3.2 Manufacturing system specification Statement 7: Information for specification of a manufacturing system can be:

a) process requirements by description of the product/products and/or in process product/products

b) set-up proposals (can be part of specification, can be part of offer)

c) technical solution constraints

d) manufacturing system properties (capacity requirements, technical availability)

e) manufacturing system constraints by description of the existing factory and building

f) regulations

g) training requirements

h) documentation requirements

i) project time constraints

Observation supporting statement 7a: Process requirements

The most essential requirement on a manufacturing system is the ability to perform the processes needed for manufacturing of specified products to be manufactured. The information for needed processes is produced during the process planning activity. In figure 52 below a process operation can be inter-preted as a single work element or as several work elements grouped into an operation.

Figure 52. General process requirements on a manufacturing system.

47

In order to create definitions of operations suitable for being specifications of functional requirements on equipment level, system design decisions have to be made, such as serial/parallel flow.

Figure 53 Example of grouping work elements into operations.

The work when grouping work elements into operations requires both process knowledge, system design knowledge and equipment knowledge. Input from equipment suppliers can be valuable, such as new possibilities to integrate sev-eral functions into the same equipment. In the connecting rod case study it was observed that the two operations OP 20 and OP 30 could be made in two set-ups using two separate fixtures and two tools in one machine tool.

In many cases multiple products/modules/parts are to be manufactured by the same manufacturing system. The requirements on the system are then an ag-gregation of all process requirements from all parts, see figure 54.

OP 20 OP 30

48

Figure 54. Aggregated requirements on manufacturing system.

By understanding the relation between functional requirements on the product and their effect on product design parameters when being changed it is possible to specify needed functionality in the manufacturing system in terms of needed variation of different process variables values. One way to express the process requirement on equipment is an indirect definition of the process by description of the wanted process output (input if necessary and possible). An alternative would be to define the required process characteristics in process-technical terms, however this has not been observed in any of the case studies.

49

0

A0

Assembly byspot welding

13103251310363

13103261310364

13102991131029921310552

Pre assemblies:Carrier frame assembly, leftpart 1310025

Carrier frame assembly, rightpart 1310026

Gusset platepart 1482735Welding

cell

A-0

Figure 55. Required function on a spot welding cell by input/output description.

The over-all total function of the welding cell was an aggregation of require-ments on performing three assembly processes. In figure 56 below the main function is broken down into three separate functions.

1

Assembly of1310025

2

Assembly of1310026

3

Assembly of1482735

A0

13103251310363

13103261310364

13102991131029921310552

Carrier frame assembly, leftpart 1310025

Carrier frame assembly, rightpart 1310026

Gusset platepart 1482735

Welding cell

Welding cell

Welding cell

Figure 56. Three independent cell functions.

50

The input parts and output parts were described in terms of material properties and geometry with acceptable tolerances and geometrical weld spot data. In complex products the product structure information containing logic has to be communicated.

Figure 57. Product information from manufacturing company.

In cases where the manufacturing company has a good understanding of the processes and the equipment to be used, a more detailed process plan might be communicated with proposed process parameter data.

51

Observation supporting statement 7b: Set-up proposal

The connecting rod case showed that one machine tool was designed to ma-chine different parts. One manufacturing functional requirement was to posi-tion the two parts for the machining operation corresponding to the process 1.2.1.3 in figure 29. Analysis of set up principle was made by aggregation of models of the parts. The two alternatives are shown in figure 58 below. The selected alternative, the right one, gives easier requirements on the fixture to be designed.

Figure 58. Two alternatives of set ups of two different connecting rods.

52

Observation supporting statement 7c: Technical solution constraints

In general it is appropriate to reuse existing manufacturing resources. This was not observed in the spot welding cell case [Mårtensson, Fagerström and Niel-sen 2002] or in the engine block line case [Appendix C]. In such a case the supplier needs information enough on the resources to be able to include them in their design. For minimisation of both operator and maintenance knowledge as well as for minimisation of spare part costs it can be relevant to constrain the make of components or the compliance to well known standards. This was made in the two cases by submitting company standard for component selec-tion.

Observation supporting statement 7d: Manufacturing system properties

For dimensioning the manufacturing system it is necessary to specify the re-quired capacity. The volume requirement from market point of view gives in-formation as input for product design decisions by constraining the process selection decisions and as capacity requirements on the manufacturing system. In all studied cases there has been a capacity requirement stated. In the spot welding cell case study [Mårtensson, Fagerström and Nielsen 2002] it was observed that there was confusion regarding the definition of capacity because of different opinion of technical availability. A conclusion is that there has to be an agreed definition of capacity and technical availability in the signed agreement between customer and supplier.

Observation supporting statement 7e: Manufacturing system constraints

Spatial constraints might be expressed as a block layout or detailed models of existing machinery and buildings. In the cab welding case no special constraints where communicated. In the engine block line case the available area was communicated.

Figure 59. Manufacturing system constraints affecting manufacturing system physical domain.

53

Observation supporting statement 7f: Regulations

The regulations around environment, health and safety valid at the place for usage have to be understood by the supplier of equipment. As one important example the 98/37/EC Machine Directive has to be fulfilled for usage within the European Union. In that case the supplier also has to understand his deliv-ery as part of a bigger system and what procedure to use for securing the com-pliance of the delivered system. References to over 50 ISO-, IEC-, EN-, DIN-, VDI- and SS-standards was found in the general specification of engine block machining case study.

Observation supporting statement 7g: Training requirements

The supplier has to understand what the users’ need of knowledge will be. This is dependent on the organisation at the user. It was found that the operators needs knowledge of safety and operation of the equipment and maintenance staff needs deeper technical knowledge.

Observation supporting statement 7h: Documentation requirements

The customer needs documentation for operation and maintenance of the equipment. It might be relevant to be able to make major changes of the equip-ment during the life of the equipment and documentation is important as input for making such changes. In both engine block line case study and the spot welding case study naming conventions for production resources at the user were communicated.

Reasoning supporting statement 7i: Project time constraints

Time plans are common project constraints. For practical installation procedure it might be relevant to specify available on-site time for installation e.g. during a holiday.

54

4.3.3 Manufacturing system physical domain The physical domain of manufacturing system contains decisions on the manu-facturing resources and the engineering work is performed by the equipment designers and is not detailed in this thesis. The equipment has to provide the function stated as product process variables. The example below shows one manufacturing system design parameter for the fulfilment of requirements on positioning two different connecting rods.

Figure 60. Manufacturing system design parameter.

Product design parameter variation requiring process variable varia-tion fulfilled by manufacturing system design parameter providing different setting

55

4.3.4 Information from supplier Statement 8: The information necessary for evaluation of a quotation and for the final delivery of a manufacturing system is:

a) time plan

b) function or process description

c) design proposal and detailed design result

d) media requirements

e) space requirements

f) training proposal

g) final documentation including spare part definition

Observation supporting statement 8a: Time plan

The time plan at the manufacturing system supplier is essential for planning the activities at the manufacturing company. In the spot welding cell case [Mårtensson, Fagerström and Nielsen 2002] a complete project plan was in-cluded in the offer.

Figure 61. Example of time plan from supplier.

Order

Process planning

Design

Purchase

Manufacturing

Installation/Testing

Transport

Installation at customer

56

Observation supporting statement 8b: Process and/or function description

The function of the offered solution can be expressed as the intended process to be executed by the equipment, similar to the process requirement as in section 4.3.2. Usually the manufacturing system supplier has prerequisites for a more detailed process description including the offered resources. Functional de-scription of offered resources can complement the process description at ten-dering phase, before detailed design of fixtures and other equipment design.

Figure 62. Two process descriptions including process and time.

3. Operator loads fixture

7. Robot welds 34 spots

30 s

85 s

57

Figure 63. Example of functional description of a fixture.

In the engine block line case the fixture design work was extensive. It was ob-served that the supplier needed a confirmation of the fixture function proposal before the detailed fixture design work started.

58

Observation supporting statement 8c: Design proposal and detailed design result

The design proposal at tendering phase cannot always be complete due to time and cost for completing a design. The manufacturing company is interested in approving the design proposals before the manufacturing i.e. after order. In the end of a project all equipment should be designed and documented by the sup-pliers. Selected and required parts of the design documents are then communi-cated to the manufacturing company. Examples of design documentation are:

- Layout describing the spatial arrangement on group/line level or on machine level.

- List of components included in the solution. Especially spare parts and wear parts.

- Drawings or models of the equipment.

- Equipment properties.

- Electrical/hydraulic/pneumatic system diagrams.

- Control system information including hardware, software, pro-grammes, and parameters.

Figure 64. Example of layout drawing including bill of material.

59

In the cab welding case [Mårtensson, Fagerström and Nielsen 2002] the design proposal including geometry and kinematics for verification of the complete work cycle was made by the supplier and communicated “live” at a project meeting.

Figure 65. Design proposal from supplier.

The final design result can be a complete description including part list with mate-rial specification and geome-try in the case where the customer buys the design.

Figure 66. Final design of in-station/ out-station.

60

Geometric simulation requires a lot of information on products, processes and manufacturing resources. Complete simulation models can be part of the final documentation so that the operational phase can be supported for changes after delivery of the system.

Figure 67. Example of models enabling geometry simulation.

61

Observation supporting statement 8d: Media services requirements

Manufacturing resources require media services from the manufacturing com-pany media system. The media services requirements help the manufacturing company to prepare the shop floor for installation. In the cab welding study [Mårtensson, Fagerström and Nielsen 2002] the supplier stated their require-ment for two separate electrical power systems. All required drop-points where marked on the offered layout, see figure 64 above.

Observation supporting statement 8e: Space requirements

The supplier knows the methods for installation and can describe requirements on space for transport and installation. The space required for the installation can be understood from the offered layout, see figure 64 above.

Observation supporting statement 8f: Training proposal

Information on training content and form is important for the manufacturing company in order to prepare the organisation for the new equipment. In the cab welding study [Mårtensson, Fagerström and Nielsen 2002] training was offered outside the original scope of delivery.

62

Observation supporting statement 8c: Final documentation

The final documentation is usually delivered after installation including late changes of the delivery. The type and amount of documentation depends heav-ily on the type of equipment. In the engine block line case study [Appendix C] the archived final documentation on one machine tool was:

Type of document Number of

sheets/ pages

Drawings 276

Part lists 164

El doc 343

Progr + param 508

Manuals 274

Other 615

Figure 68. Archived information on one machine tool.

For easier use in operation, certain maintenance instructions for operators have been created. The information references the original final documentation from the supplier, see figure 69 below.

Figure 69. Example of maintenance points on machine layout and corresponding main-tenance instruction.

63

5 DISCUSSION AND CONCLUSIONS

This chapter discusses the results in chapter 4 by relating them to the hypothe-sis and its validity. Furthermore points for further research are proposed.

5.1 Discussion The research question was defined as “What method could be used for im-proved development of manufacturing systems?”. The research work has been divided into two manufacturing system development areas:

- specification of the manufacturing system

- communication between the manufacturing company and the manufac-turing system supplier during development

The first part of the hypothesis deals with the specification of the manufactur-ing system. The second part of the hypothesis deals with the communication.

5.1.1 Discussion on hypothesis One major obstacle in specifications of a manufacturing system is the different life span of the product to be produced and the manufacturing system to be used for manufacturing the product. Products life time might be a couple of years, where machinery has a technical life time reaching far longer, given that proper maintenance is being performed.

The solution to the problem is stated in the hypothesis where the aim is to fore see the impact of product changes on the process. By analysing the product design parameters relation to the product function a proper basis for process decisions, and in the end, a proper basis for specification of equipment can be stated. Rather than seeing the product models as static models they should be seen as dynamic models that can evolve in certain ways for certain reasons. There is a risk that new technologies give totally new product concepts and design. In such cases the proposed method does not give any particular support.

64

The strategy expressed in chapter 4 is to analyse the logic between product design, process planning and equipment design. By understanding what deci-sions are crucial and the basis for those decisions one can make sound deci-sions.

In the rear axle case study [Mårtensson 2000] the problem with product changes affecting the clamping conditions showed the risk for couplings be-tween functional requirements on the product and requirements from the proc-ess. When the products were changed, the clamping solutions had to be changed. The case study on pinion manufacturing gave input for formulation of the in process requirements on the in process product, see appended paper B.

The conclusion was that the analysis of the product to be produced has to start with understanding the customer needs and clear definition of need for variance in the product. The validation case study on connecting rods showed the same needs and the experienced people involved in the project seemed to think in terms of future possible changes driven by changed customer needs and con-straints from environmental regulations.

The described logic in chapter 4 connects product function with requirements on the equipment using product design parameters and process variables.

The design of the manufacturing system and its components are usually per-formed by specialised suppliers. The suppliers need information from the manufacturing company in order to select/adapt/design solutions. The comput-erised development tools at the suppliers represent information coming from the manufacturing company, especially product data. One time consuming activity is the creation of already existing information inside the manufacturing system supplier. The corresponding process takes place in the reverse direction.

The requirements on the information exchange expressed in chapter 4.3 and 4.4 have lead to the strategy of using international information standards for enabling computer based information exchange.

Based on studies of project information exchange in the cab welding case study a solution based on STEP AP 214 ARM was developed and tested [Mårtens-son, Fagerström and Nielsen 2002]. Additional requirements and necessary changes where made during the validation case (engine block machining). One major limitation of the test implementation was that factory constraints, tech-nology choice constraints and process were not covered. The reason for this was that there were no internal systems for holding such information in a struc-tured way at the manufacturing company.

Two problems with the proposed solution concept have been identified:

65

The first problem is the establishment of the standards in terms of implementa-tion and usage. If there is only one telephone the telephone is useless even though it works. This was the major reason to choose the STEP AP 214 as the basis for communication. It is a quite well established standard within the product design area within automotive industry, particularly in Europe. How-ever the use for process planning and equipment design it is not well estab-lished. Extensive usage relies on the possibility to use commercial import- and export functions developed by information system suppliers.

The second problem is the feasibility of the information standards to truly rep-resent the information to be communicated. In the documentation studied in the two case studies concerning information exchange during development of manufacturing systems not much information on possible future product changes was expressed. However in the personal project meetings and other non documented contacts requirements on needed variation has been ex-changed. The developed solution does not cover communication of such ad-vanced information. A more solid evaluation of the usage of standardised rep-resentation of the information is published in [Nielsen 2003].

The recently published standard STEP AP 240 for process planning of ma-chined parts is an interesting complement/alternative to STEP AP 214 for de-velopment of machining systems. There is a need to harmonise the two stan-dards and there is a need to cover other types of processes than machining also within machining systems.

66

5.2 Scientific contributions • Three general types of three domain product system structures have

been defined. These support high level decision making for product planning and can also be detailed during development. By using obser-vations from the case studies Suhs [1990] three domain model (FR, DP, PV) has been illustrated and the general types have been defined, see statement 1,2,3 resulting in figure 24.

• Six backgrounds to requirements on a process have given six groups of processes. The processes are grouped from a functional point of view rather than an enumeration of processes. The groups help the process planner to ask the right questions when stating the requirements on production equipment, see statement 4.

• Definition of “in process functional requirements” and “in process de-sign parameters” as a certain view. Axiomatic Design does not explic-itly handle the internal requirements and does therefore not give any support for production adaptation of products. The proposed extra view can be used on top of the original domains for this purpose, see figure 31.

• Process operational sequence implication on information content, fig-ure 43 and 44 and additions related to in process view. Sequence deci-sion impacts number of functional requirements as well as difficulty.

• A description of general process implications on product design based on observed examples, see statement 6.

• Manufacturing system specification concept based on manufacturing functional requirements as “perform process operation”. The process operations are grouped process variables acting on design parameters or in process design parameters, see figure 52 and figure 54.

• Illustration of relation between design parameter, process variable and manufacturing system design parameter, see figure 60.

• Contribution to the evaluation of usage of STEP AP 214 for manufac-turing system design by providing a description of the needed informa-tion to be communicated between manufacturing company and manu-facturing equipment suppliers, see appendix C.

67

5.3 Conclusions There is logic between product design, process planning and manufacturing system design. The logic can be described and used at least internally within a company. As a complement to static product models the understanding of the background of the product design can help the process planner, production engineer and possibly the manufacturing system supplier to have a more com-plete understanding of the requirements on a manufacturing system. By having a more complete understanding of the requirements a more appropriate manu-facturing system can be designed. The use of information exchange solutions can speed up the work and also increase the precision of engineering decisions and by that raise the quality of the manufacturing system. Implementations based on international standards are possible. The area of international infor-mation standards is evolving and competing standards have to be harmonised or redefined.

5.4 Future research First of all the developed method and communication concept should be tested in ongoing projects within automotive industry.

There is a need to perform project work completely using what has been shown as fragments in this thesis. That would give an understanding of applicability and usefulness that is more solid than what has been possible to do within this research project. It will always be difficult to evaluate the bottom line results of different approaches by the nature of the subject. A project is seldom per-formed more than once and by that it is not possible to compare the result to any alternative result.

Secondly it would be interesting to study the applicability in other industries and to find possible modifications and extensions.

There is a need to develop standardised work methods and guides for imple-mentation that are accepted in industry. Initial tests of standard based commu-nication indicates a rather big barrier and points out needs of means for reusing already existing information more or less automatically.

There is a huge potential in maximising quality and minimising effort by con-scious statement of requirements (product design parameters) and conscious selection of processes. There is therefore of utmost importance that research projects study processes’ affect on product properties/function, ideally to the level of mathematical models. Analysis of the process chains across company

68

borders is necessary to reach the maximum of quality using a minimum of ef-fort.

69

REFERENCES

Almström P., Mårtensson P. (2001) “Functional coupling in manufacturing systems and its implications”, Proceedings of International CIRP Design Semi-nar – Design in the New Economy, Stockholm, Sweden

Alsterman H. (2001) “Towards hyper-flexible assembly automation – devel-opment of a hyper-flexible assembly system concept for light mechanical as-sembly”, Licentiate thesis, Division of assembly systems, Department of Pro-duction Engineering, KTH, Sweden, TRITA-IIP-01-18, ISSN 1650-1888

Andersson-Stråberg A. et al (2005) ”Facts about the Swedish economy 2005”, Confederation of Swedish Enterprise, Davidsons Tryckeri, Växjö, Sweden, ISSN 0280-3364

Askin, R.G., Standridge C.R. (1993) “Modelling and analysis of manufacturing systems”, John Wiley & Sons, Inc, Singapore Bley H. and Franke C. “Integration of Product Design and Assembly Planning in the Digital Factory”, Annals of CIRP, Vol 53, p. 25 CIRP (2004) “Dictionary of Production Engineering III”, Springer-Verlag Ber-lin Heidelberg, Germany, ISBN 3-540-20555-1

DESINA (2001) “Detailed cost analysis of a machining center”, project docu-ment, available on http://www.desina.de/english/desina_homepage_e.htm, checked 2004-04-06

Ekdahl H. (2003) “Frihandel och investeringar ger välstånd”, report, Confede-ration of Swedish Enterprise, Sweden

Eriksson G. (1996) ”Modularity – the basis for product and factory reengineer-ing, Annals of CIRP, Vol 45

EUPASS (2006) ”Evolvable Ultra-Precision Assembly System, 6th Frame-work”, Internal project information, Courtesy TQC Ltd, UK

Evbuomwan N.F.O, Sivaloganathan S., Jebb A. (1996). ”A survey of design philosophies, models, methods and systems”, Proceedings of the Institution of Mechanical Engineers. v 210, p. 301-319, 1996

Handfield R.B., Melnyk S.A (1998) “The scientific theory-building process: a primer using the case of TQM”, Journal of Operations Management 16, pp 321-339, Elsevier Science B.V.

70

Habel P. (2003) “PDTnet implementation guide v 1.6”, Project report, avail-able on www.pdtnet.org, checked 2004-03-17

ISO (2001) “ISO 10303-214:2001(E), Industrial automation systems and inte-gration – product data representation and exchange – part 214: Application protocol: Core data for automotive mechanical design process”, ISO

ISO (2004) “ISO/DIS 10303-240:2004(E), Product data representation and exchange: Application protocol: Process plans for machined products”, ISO

Johansson L. (2003) ”Kommunikation av produkt-, process- och resursdata baserat på Internet och standards”, Final project report, IVF, Sweden

Johansson M. (2001) “Information management for manufacturing system de-velopment”, Doctoral thesis, Division of computer systems for design and manufacturing, Department of production engineering, KTH, Sweden, TRITA-IIP-01-04, ISSN 1650-1888

Jovane F., Koren Y., Boër C.R. (2003) ”Present and future of flexible automa-tion: Towards new paradigms”, Keynote paper, Annals of CIRP Vol 52/2.

Krause F. L-. et al (2003) “Computer aided conceptual design”, Proceedings of the 36th CIRP international seminar on manufacturing systems, Saarbrücken, Germany

Mayer R., Menzel C., Painter M., deWitte P., Blinn P., Perakath B. (1995) ”Information integration for concurrent engineering IDEF3 process desription capture method report”, interim technical report, Knowledge based systems Inc, Texas, USA

Ministry of finance (2005) “Swedish economy”, Appendix 2 to Budget Bill 2006, PROP 2005/06:1, Stockholm, 2005

Mårtensson P., Tate D. (1998), “Reuse of Design Rationale in Cellular Manu-facturing Systems: Theory and Application”, Proceedings of the 2nd Conference on Engineering Design and Automation, ED&A ´98, HI, USA

Mårtensson P. (2000), “Manufacturing subsystem design decisions”, Proceed-ings of the 33rd CIRP International Seminar on Manufacturing Systems, 5-7 June 2000, Stockholm, Sweden

Mårtensson P. (2003), “Communication for development of machining systems - method and implementation based on STEP and XML”, not yet published

Mårtensson P., Du J., (2001), “Crank shaft manufacturing simulation”, project report, Lund University, Sweden

71

Mårtensson P., Fagerström J. (2000) “Product Function Independent Features in Axiomatic Design”, Proceedings of the First International Conference on Axiomatic Design, ICAD 2000, June 21 – 23 June 2000, MA, USA

Mårtensson P., Nielsen J., Fagerstöm J. (2002) “Co-operative digital projecting of manufacturing equipment – method and implementation based on STEP and XML”, Proceedings of the 33rd International Symposium on Robotics, Stock-holm, Sweden

Nielsen J., Kjellberg T. (2000) “The ISO 10303-214 process model as a core for a process-planning tool”, International CIRP Design Seminar, Haifa, Israel

Nielsen J. (2003) “Information modelling of manufacturing resources - infor-mation requirements for process planning in a concurrent engineering envi-ronment”, Doctoral thesis, Department of production engineering, KTH, Swe-den, TRITA-IIP-03-09, ISSN 1650-1888

NIST (1993) “Integration definition for function modelling (IDEF0)”, Draft Federal Information Processing Standards Publication 183, 1993, USA

Onori M., Barata de Oliveira J., Frei R. (2006) ”Evolvable Assembly and Ex-ploiting Emergent Behaviour”, To be published in proceedings of ISIE2006, Montréal, Canada

Petersson (2000) “Process Efficiency and Capacity Flexibility – Developing a Support Tool for Capacity Decisions in Manual Assembly Systems”, Doctoral thesis, Production Systems, Department of Mechanical Engineering, Linköpings universitet, Sweden

PricewaterhouseCoopers´ (2002) ”Automotive Sector Insights – Analysis and opinions on Merger Acquisition Activity 2000/2001”, company report

Randell L. (2002) “On discrete-event simulation and integration in the manu-facturing system development process”, Doctoral thesis, Division of robotics, Department of mechanical engineering, Lunds universitet, CODEN: LUTMDN/(TMMV-1054)/1-165/2002, ISBN 91-628-5319-8

Rech R., Klein L. (2001) “STEP-Integrationsplattform Digitale Fabrik”, Pro-duktDatenJournal Nr 2, pp 19-22, ProSTEP Association Gmbh, Darmstadt

Rech R., Klein L., Randis R., Baltramaitis T., Nargelas V. (2004) „Integrating Distributed Applications on the Basis of STEP Data Models, Final Report

Robertson D., Ulrich K. (1998) “Platform product development”, version 2, Sloan Management review, p 19-31

SCB (2005) “Investments in building and machinery in 2005, according to the May survey” NV 10 SM 0501

72

Sjöström S. (1991) ”The modular system in truck manufacturing”, The Saab-Scania griffin, 1990/1991

Sohlenius G., Sandell P. (1975) “Prodevent – Coordinated customer directed manufacturing” in Swedish, IVF-resultat 76604, april 1976, ISBN 91-524-0285-1, Sveriges Mekanförbund, Stockholm, Sweden

Sohlenius, G. (1980) The manufacturing system nomenclature, structure and content, Hierarchy of Terminology for Manufacturing, CIRP-project, Beograd

Sohlenius G. (2001) “Engineering science – qualities and characteristics”, pre-pared for CIRP working group meeting, August, Nancy

Sohlenius G. (2002) “Engineering Science – Qualities and Characteristics”, International Journal of Production Engineering and Computers, Vol 4, pp 9-12

Sohlenius G. (2004) “Systemic nature of the innovation process in industrial production - our engine of welfare”, Forthcoming doctoral thesis, Tampere University of Technology, Finland

Suh N.P. (1990) “The principles of design”, Oxford university press Inc., New York, USA, ISBN 0-19-504345-6

Svensson I., Svensson I. (2005) ”Karlebo Gjuteriteknisk Handbok”, NRS Tryckeri, Huskvarna, Sweden, ISBN 91-631-6158-3, p 456, p 457, p 460

Tseng M., Xuehong D. (1998) “Design by customers for mass customization products”, Annals of CIRP Vol 47/1 p. 103

Tseng M., Kjellberg T., Lu S. (2003) ”Design in the new e-commerce era”, Keynote paper, Annals of CIRP Vol 52/2

Ullrich K., (1995) “The role of product architecture in the manufacturing firm”, Research policy Vol 24, pp 419-440

W3C – world wide web concortium (2001), www.w3.org, checked 2004-03-18

Wozny M. J. (1989) “Toward a theory and methodology of design”, Proceed-ings of the 3rd international conference on foundations of computer aided proc-ess design, USA

Xiao (2002) “Effect of rough surface anisotropy on friction in gears”, Licenti-ate thesis, Department of product and production development, Chalmers, Sweden

Ålund P., PDTnet consortium (2001) “PDTnet – Leveraging product data inte-gration using internet technology”, AutoTech, Detroit, USA

APPENDED PAPER A: MANUFACTURING SUBSYSTEM DESIGN DECISIONS Proceedings of the 33rd CIRP International Seminar on Manufacturing Systems, 5-7 June 2000, Stockholm, Sweden.

Published in CIRP Journal of Manufacturing Systems

Volume 31, 2001, No. 1

APPENDED PAPER B: PRODUCT FUNCTION INDEPENDENT FEATURES IN AXIOMATIC DESIGN

Co-authored with Jonas Fagerström.

Proceedings of the First International Conference on Axiomatic Design, ICAD 2000, June 21 – 23 June 2000, MA, USA

APPENDED PAPER C: FUNCTIONAL COUPLING IN MANUFACTURING SYSTEMS AND ITS IMPLICATIONS

Co-authored with Peter Almström

Proceedings of International CIRP Design Seminar – Design in the New Econ-omy, 6 – 8 June 2001, Stockholm, Sweden

Published (compact version), A categorization of functional couplings in manufacturing systems, in Journal of Engineering Manufacture, Volume 216, No 4, 2002

APPENDED PAPER D: CO-OPERATIVE DIGITAL PROJECTING OF MANUFACTURING EQUIPMENT - METHOD AND IMPLEMENTATION BASED ON STEP AND XML

Co-authored with Jonas Fagerström and Johan Nielsen.

Proceedings of the 33rd ISR, International Symposium on Robotics,

October 7 – 11, 2002, Stockholm, Sweden

A1

0

A0

Function name

A-0

Input

Control

Output

CallMechanism

1

2

3

4A4

A0

1

2

3

A4

More general

More detailed

APPENDIX A: IDEF0 – FUNCTION MODELLING METHOD Integration Definition for function modelling, IDEF0, is used to represent func-tions, activities or processes within the modelled system or subject area. The over-all system function is described in an A-0 top level context diagram along with model purpose and viewpoint. The top level child diagram is always the A0 diagram. Boxes arranged diagonally are numbered in order from left to right (chronological order) and off diagonal boxes are numbered counter clockwise with higher number than the diagonal boxes. However, functions represented by boxes may be active simultaneously and one should not think that the sequence has a meaning.

IDEF0 is based on that functions can be described and communicated in a hierarchy going from general level to more detailed level Figure A1. IDEF0 modelling principles, adopted and modified from [NIST 1993]

B1

APPENDIX B: IDEF3 – PROCESS DESCRIPTION CAPTURE METHOD The IDEF3 modelling method [Mayer et al 1995] is used to describe processes and object states. The process description notion is not used in this thesis. The Object State Transition Network Diagram (OSTN diagram) is used in this the-sis to capture information on the process and the operand from objects’ point of view. In figure B1 below the used symbols are presented.

UOB/ P

a

A

Kind object a

B

Object a in state Afollowed by anobject b in state B(a and b usually the same object)

a:AKind object a in state A

A B

Transition schematic with UOB(units of behaviour) referent.Process P is involved intransition from state A to state B

Figure B1. Object schematic symbols in IDEF3 and a referent symbol

C1

APPENDIX C: CO-OPERATIVE MODEL BASED MANUFACTURING SYSTEM DESIGN

This appendix describes the scenario and solutions for communication between the manufacturing company and the equipment suppliers as developed and tested within the PDTnet project.

Information requirements This appendix is mainly based on two case studies on project communication between end user and equipment supplier. One case study concerned the devel-opment of a cab spot welding cell and one case study concerned the develop-ment of an engine block machining line.

Some information requirements has been stated based on decisions described in section 4 in this thesis. Commercial information requirements are not included.

Communication solution This section is based on the cab welding case study and the engine block ma-chining case study performed as parts of the PDTnet project, product data tech-nology and communication in an OEM and supplier network. The PDTnet target scenario is shown in figure C1 below.

Figure C1. PDTnet target scenario [Ålund, PDTnet consortium 2001].

Product Data Communicationbased on

Harmonised Processesand Standards:

STEP AP 214XMLOFTP

Internet ProtocolsENX (GNX)

PDMSystem

PDMSystem

PDMSystem

PDMSystem

NeutralWeb

Client

PDMTool

C2

PDTnet schema The PDTnet schema [Habel 2003] is a defined subset from the product data standard ISO 10303-214. The main reason for the development of the PDTnet schema was to enable communication via the language XML, extensible markup language. XML schemas express shared vocabularies and provide means for defining structure and content of XML documents. XML schema is an approved W3C Recommendation [W3C 2001].

PDTnet schema was initially defined for product design needs where manufac-turing companies communicates with suppliers. Based on information from the research described in this thesis and in [Nielsen 2003], the PDTnet schema was extended with process information entities. The PDTnet schema covers:

- External reference mechanism (documents) - Part property

- Part definition structure - Product management data

- Effectivity - Work management

- Specification control - Process plan

Demonstrator implementations Three demonstrators where developed in order to prove the communication concept. The first demonstrator was a feasibility test, the second an implemen-tation where existing systems were adapted and the third was a test where ex-tensions for covering needs for machining systems were made.

Export of information from manufacturing company

The export function was limited to product data export covering product struc-ture with references to attached documents (CAD-models, standard documents, material specifications etc) and the developed solution was made extendable. Ideally the export function should cover needs for communication of factory constraints and technology choice constraints as well as process planning in-formation. The export function consists of four sub-functions, see figure C2.

C3

Figure C2. Implemented manufacturing company information export function.

The demonstrator used existing information sources where part information gathering function had to be developed and where mapping between the source schema and target schemas had to be developed.

1

Specify product

2

Gather partinformation

3

Translate tostandard format

4

Pack and sendinformation

Productstructuresystem

Partinformation

systems

STEPprocessor

Target schemaTarget file format

Neededinformationtypes

Conditions forspecification

Recieverinformation

Dataexchange

system(ENGDAT)

ENGDATpackage

Product data file

Part informationdocuments

Product data

C4

Import of product information into the manufacturing system supplier

The requirements on the import function were limited to product data import. Ideally factory constraints, technology choice constraints and process planning information should be covered as well.

Figure C3. Implemented manufacturing system supplier information import function.

The demonstrator used a system specific non-commercial existing STEP import system for PDM schema, part 21 files. Part information document import was not made automatic within the developed demonstrator and the import had to be performed manually for the test case illustrations.

1

Recieve andunpack

information

2

Translate tospecific formats

Dataexchange

system(ENGDAT)

STEP importsystem

ENGDATpackage

3

Import and storeinformation

Documents

Informationready for usage

Internalsystems

Productdata file

C5

Export of manufacturing system supplier information

The export function utilised the potential in representing the relation between products and manufacturing resources by utilisation of process plan informa-tion. The mapping between the native XML file and the PDTnet XML file was developed by the project.

Figure C4. Implemented manufacturing system supplier information export function.

Viewing and editing product, process and resource information

The import of information from the manufacturing system supplier into the manufacturing company could not be realised for practical reasons. The neutral web client was used for the demonstrator to verify the final result of the chain of information.

1

Extractinformation frominternal system

2

Translate tostandard format

Internalsystem

functions

XSLTprocessor 3

Pack and sendinformation

Documents

PDTnet schemaXML file

Internalsystems

Product, process and resource information

Native XML file

ENGDATpackage

C6

Demonstrator architectures Two concepts where developed, depending on different pre-conditions. One concept is based on the assumption that the involved users have a PDM-like system, see figure C5 below. The other concept is based on the assumption that some users have no internal system for information management, see figure C6 below.

ProductStructuresystem

ModelArchivesystem

STEPprocessor

ViewerEditor

Dataexchange

tool

File server

Dataexchange

tool

XSLT

Developmentsystem

STEP-importer

File server


Manufacturingsystem supplierPDM

schemapart 21

file

PDTnetschemaXML file

Figure C5. Concept for users with own PDM-like system.

ProductStructuresystem

ModelArchivesystem

STEPprocessor

ViewerEditor

Dataexchange

tool

File server

Webclient

File serverSOAP

interfacePDTnetclient

installation



PDTnetschemaXML file

Figure C6. Concept for users without own PDM-like system.

C7

Development process illustration This section illustrates the communication during a manufacturing system de-velopment process using information from the cab case study from the systems developed and used within demonstrator 2.

Export of information from manufacturing company

The cab is a modular component requiring conditions to be set for specification of the components. For this reason a product specification and creation of product structure (BOM) has to be performed, see figure C7.

Figure C7. Example of product specification at manufacturing company.

From the structure system an EXPRESS file expressing information in com-pany specific schema is created. Using the mapping function a new file with the needed schema and file format can be created. The source file and a result-ing file are shown in figure C8 below. In the example the target schema was set to PDM schema, part 21. All information within the source file is not covered by the PDM schema e.g. multiple languages. The source system has no infor-mation on assembly sequence, and therefore some additional parts are de-scribed compared to the parts for assembly in figure 55 above. The additional parts are assembled in pre-assembly operations being performed before the assembly of the main parts 1310325 and 1310363.

C8

Figure C8. Example of source file extract and PDM schema file extract.

Import of information into the manufacturing system supplier

The product structure information is imported using a PDM schema file import function and the geometric files are imported using a system specific direct converter. Other documents can be imported if needed.

Figure C9. Example of imported product structure and geometric information at manufacturing system supplier.

ISO-10303-21;HEADER;FILE_DESCRIPTION(('Product structure… '),'2;1');FILE_NAME('Cabin_020807.stp', …);FILE_SCHEMA((…));ENDSEC;

DATA;#1…#519=MULTILANGUAGESTRING((#520,#521,#522,#523,#524,#525));#520=STRINGWITHLANGUAGE('CARRIER FRAME ASSY LH',#20);#521=STRINGWITHLANGUAGE('ARMACION IZQUIERDA',#26);#522=STRINGWITHLANGUAGE('CHASSIS PORTEUR CPL GAUCHE',#22);#523=STRINGWITHLANGUAGE('MONTAGEPLAAT',#23);#524=STRINGWITHLANGUAGE('ARMACAO ESQUERDA',#24);#525=STRINGWITHLANGUAGE('MONTERINGSPL\X\C5T D\X\D6RR KPL V\X\C4',#21);#2379=PHYSICALPART(#519,'0000001310025',$,#30,$,.S.,#325,'041240','0',.U.,.SSS.,(),(#2538),#2041,.ST.,#140,$,$,$,#69,#140,(#2ENDSEC;END-ISO-10303-21;

ISO-10303-21;HEADER;FILE_DESCRIPTION(('Door left'),'2;1');FILE_NAME('door_left_PDMexport.stp', '2002-08-14T13:35:13',...);FILE_SCHEMA(('PDM_SCHEMA {1.1}'));ENDSEC;

DATA;#570=PRODUCT('0000000815551', 'SQUARE WELD NUT, UNTAP#573=PRODUCT('0000000815552', 'SQUARE WELD NUT,UNTAP#574=PRODUCT_RELATED_PRODUCT_CATEGORY('assembly', #586=PRODUCT('0000001310025', 'CARRIER FRAME ASSY LH', #603=PRODUCT('0000001310220', 'REINFORCEMENT', $, (#905)#614=PRODUCT('0000001310325', 'DOOR CARRIER REINFORCE#615=PRODUCT('0000001310363', 'DOOR CARRIER CROSSMEMENDSEC;END-ISO-10303-21;

C9

Process planning and equipment design

The development system at the manufacturing system supplier supports process planning and equipment design activities. The process discription shown in C10 below is equivalent to the process description in figure 62 above.

Figure C10. Process planning support at the manufacturing system supplier.

C10

Figure C11. Example of equipment design and evaluation at manufacturing system supplier.

Export of information from the supplier

The created process and equipment structure information can be exported as a PDTnet schema XML file by using an export function developed by the pro-ject. Geometric files are collected and placed in the same directory as the struc-ture file.

Figure C12. PDTnet schema export function.

C11

Viewing and editing product, process and resource information

The neutral web client solution can be used for viewing and editing the product, process and resource information expressed as PDTnet schema, XML-file. In figure C13 below the process structure is broken down to task level and the assigned resource, the inputs and the output to/from the operation is shown. In figure C14 an example of resource structure is shown.

Figure C13. Process structure and related information in neutral web client, cab case

C12

Figure C14. Example of manufacturing resource structure.

Design Decisions and Co-operative Development of ...10385/FULLTEXT01.pdf · design and the...

Documents

Transcript of Design Decisions and Co-operative Development of ...10385/FULLTEXT01.pdf · design and the...