The Research And System Development Of Product Configuration By Combining OO And Constraint Reasoning

Nowadays the competition between companies is increasingly intense。The companies now face numerous challenges to succeed in business and make profit. On the other hand, the more emphasis on personalized awareness implies the increase of the demand for personalized products. Therefore, How to maximally satisfy this kind of needs from customs turns out to be the key problem of winning the challenges for them. To solve this problem the global business community has begun a new attempt to implement mass customization. And thereby product configuration, one of the key enablers of mass customization, becomes a hot research field of academic community. In these years domestic and foreign scholars have deeply investigated product configuration methods and obtained abundant research results which make product configuration systems to be applied in many field such as electronics,computer,telecommunication.In this paper, the forming and evolving process of product configuration,the application under the mass customization paradigm as well as the research overview was discussed in detail. In order to implement configuration systems efficiently and solve configuration problems effectively, a new approach combining object-oriented modeling technology and constraint reasoning was put forward. In presented approach, firstly a reusable and sharable configuration model was established to represent domain-specific knowledge by using the object-oriented modeling approach. Then the configuration model was mapped to a Dynamic Constraint Satisfaction Problem (DCSP) and finally the configuration problem was solved efficiently through the use of constraint reasoning. To model a three-layered architecture (meta model,model,object)was put forward to enhance model reusability. When mapping from the configuration model to DCSP mapping rule of Cardinality problem,structured relationship,configuration constraints were researched and a package of mapping rules was developed. For solving the DCSP an extended backtracking algorithm was designed and its inference process was demonstrated.HIGER BUS Company Limited is a notable bus manufacturer in China.。Up to June 2007, the company's HIGER BUS has produced and sold over 70,000 buses, which has a huge guaranteed market share. In this paper, a configurable HIGER BUS was presented to demonstrate our configuration method and illustrate the effectiveness of the presented algorithms. Finally the OO-DCSP-based bus configurator shell was developed, which was coded with JAVA programming language.