Platform based evolutionary product customization

Abstract

This paper is concerned with the development of a methodology for customizing product designs from a product platform using genetic algorithm to satisfy customers' requirements. The structure of the product platform was modeled by an AND/OR/RL tree. Each node in the tree denotes a module, a component or a variable. The optimal product design includes the structural design and its relevant parameters. Two-layer evolutionary algorithm was employed for identifying the optimal structure and its parameters. A case study on the worm gear reducer was given to illustrate the framework and the proof-of-the-concept prototype system while the framework must be further substantiated through more comprehensive case studies.