Improved Genetic Programming for Process Planning

Abstract

The traditional manufacturing system research literature generally assumed that there was only one feasible process plan for each job. This implied that there was no flexibility considered in the process plan. But, in the modern manufacturing system, most jobs may have a large number of flexible process plans. So, flexible process plans selection in a manufacturing environment has become a crucial problem. In this chapter, a new method using an evolutionary algorithm, called Genetic Programming (GP), is presented to optimize flexible process planning. The flexible process plans and the mathematical model of flexible process planning have been described, and a network representation is adopted to describe the flexibility of process plans. To satisfy GP, it is very important to convert the network to a tree. The efficient genetic representations and operator schemes also have been considered. Case studies have been used to test the algorithm, and the comparison has been made for this approach and Genetic Algorithm (GA), which is another popular evolutionary approach to indicate the adaptability and superiority of the GP-based approach. The experimental results show that the proposed method is promising and very effective in the optimization research of flexible process planning.