An approach to acquiring configuration rules using rough set theory

Configuration design is one of the most important contents in the design for product family, with which the designers could customize product for customer according to his specific requirements. The conventional approach to configure product is based on configuration rules, which reflect the mapping relations between customer requirements and design variables. But it is not easy to acquire configuration rules, which generally depend on design expert himself to do it. And it would take plenty of time for design experts to establish rules database. This paper presents a new approach to acquiring configuration rules based on the idea of data mining using rough set theory from large numbers of historical design data instead of design expert experience. It can help not only design experts but also novice designers to analyze the dependency between customer requirements and design parameters, and to acquire configuration rules hidden in historical data easily.