Project 5: Configuration Rule Management

​Project Leader: Johan Malmqvist (Research group: System Engineering & PLM)

Academic staff

RG: Systems Engineering & PLM (SE)
Johan Malmqvist
Anna Tidstam, PhD Student from year 4

RG: Geometry and Motion Planning (GMP)
Fredrik Ekstedt

RG: Flexible Automation (FA)
Knut Åkesson
Alexey Voronov, PhD Student

Industrial partners

Volvo Cars AB
Bertil Turesson

Saab Automobile
Anders Claesson
Joakim Olsson
Johan Tärbo

Volvo Lastvagnar
Samir Mesihovic
Camilla Nordström
Anna Tidstam, Industrial PhD Student year 3 ´

FCC
Fredrik Ekstedt
Oxana Sachenkova
Suvash Thapliya

Background

A configurable product platform is an efficient way to handle a wide range of product variation and offering and at the same time utilize a limited number of parts and components. To manage this in the different processes in the company, the product documentation is managed through a PDM system. However, when the variation of the product offering is increasing over time – including merger and acquisitions of companies – the environment for the engineer, when doing product changes, becomes more and more complex despite the support from the PDM system. Therefore, in the PDM system there is a need to organize the data into manageable sub-units in order to reduce the complexity for the engineer. Different solutions for the organization of sub-units will have different impact on the engineering documentation efficiency, and an optimization is needed. KOLA-logic is the way product definition information is organized and managed in the Volvo Group truck business’s main processes, Product Development, Sales-to-Order, Order-to-Delivery, and Delivery-to-Repurchase. It has been built on the basis of experiences (“know-how”) step by step during many years. A first attempt to create a complexity indicator has been developed.

Goal

The purpose of this project was to create tools and strategies in order to understand what drives documentation complexity, and how to reduce it in order to make engineering documentation more efficient. The project addressed the following research questions:
  • RQ 5:1
    What is a good set of configuration rules and how should it be structured?
  • RQ 5:2
    What is an efficient process for formulating/ using/maintaining the configuration rules?
  • RQ 5:3
    How can the documentation of the configuration rules be improved?
  • RQ 5:4
    How can the system support for formulating/ using/maintaining configuration rules be improved?
  • RQ 5:5
    How can Binary Decision Diagrams support and facilitate the utilization and maintenance of the variant families and their restrictions?

Project realization

The project was carried out in close cooperation with our industrial partners, and also involved people from these companies and involving international mathematical experts in the area of PDM. The project used the Volvo PDM, KOLA-Logic as the application to study.

Summary of results

The project has mainly resulted in new information models, processes, algorithms and computer tools for configuration rule authoring, which have been tested and implemented in three automotive companies. RQ5:1-5:3 mainly addressed what is a suitable information model for automotive configuration rules, and what is an efficient process for authoring such rules. In [5:2 and 5:3] such information models and processes are presented. RQ5:4-5:5 mainly addressed suitable computer support for authoring, validating and computing configuration rule sets. In [5:4], an algorithm and tools for computing validation configurations was developed. In [5:1] a visualization tool that supports authoring and validation is presented.

Implementations

CONFIRM – Tool for computing and analysis of valid partial configurations


Image: Configuration Variants
Short description/results:
Several methods to compute valid partial configurations were developed and evaluated using real data from Volvo 3P and Saab Automobile [5:4 and 5:8].
 
Implementation:
Industrial data were used to evaluate the proposed methods. This evaluation received positive feedback from Volvo 3P. A pilot implementation project was started by Volvo 3P and Volvo IT to introduce the methods into Volvo’s environment. A company (Confirmlogic AB) has been started to provide a ‘beyondthe- research-prototype’ implementation of the proposed methods. Up to now, the company has sold the tool to Volvo 3P.

Publication and Presentation Activity

5.1 Tidstam, A., Bligård, L.-O, Voronov, A., Åkesson, K. , Malmqvist, J., 2012, “Development of Visualization Support for Validation of Vehicle Configuration Rules”, accepted for Proc. of TMCE2012, Karlsruhe, Germany.
 
5.2 Tidstam, A., Malmqvist, J., 2010, “Information Modelling for Automotive Configuration”, Proc. of NordDesign 2010, Gothenburg, Sweden.
 
5.3 Tidstam, A., Malmqvist, J., 2011, “Authoring and Verification of Vehicle Configuration Rules”, Proc. of Product Lifecycle Management (PLM11) Conference, Eindhoven, The Netherlands.
 
5.4 Voronov, A., Åkesson, K., Ekstedt,F., 2011 “Enumeration of valid partial configurations”, Proc. of IJCAI 20011 - Workshop on Configuration IJCAI 2011, Barcelona, Spain.

Theses

5.5 Hu, Z., 2010,”Analysis and Presentation of Combinatorics in Product Configuration”, Master Thesis, Chalmers University of Technology, Gothenburg, Sweden.
 
5.6 Sachenkova, A., Thapaliya, S. K., 2011, “Using CSP solvers for Partial Configuration in Automotive Configuration Problems”, Master Thesis, Chalmers University of Technology, Gothenburg, Sweden.
 
5.7 Tidstam, A., 2012, “Developing Vehicle Configuration Rules”, Licentiate thesis, Department of Product and Production Development, Chalmers University of Technology, Gothenburg, Sweden.
 
5.8 Voronov, A., 2010, “Using Formal Methods for Product and Production Development – Industrial Applications for Boolean Satisfiability Solvers”, Licentiate Thesis, Department of Signals and Systems, Chalmers University of Technology, Gothenburg, Sweden.

Published: Wed 28 May 2014.