Optimization of Genetic Algorithm Parameters for Multi-channel Manufacturing Systems by Taguchi Method

Abstract

Abstract. An important issue in multi-channel manufacturing (MCM) design is the channel formation process. In this study, the control parameters that affect the performance of genetic algorithms (GAs) developed to solve channel formation problem, are examined and the optimum values of such parameters are explored using Taguchi method. Two types of problems were taken into account in terms of machines, parts, and channels. Experimental results show that the performance of a GA significantly dependent on the levels of the design factors for the problem being solved. The results also show that Taguchi method is a powerful approach for identifying design factors suitable for the GA comparing to time consuming and possibly impractical trial-error tests.