A hybrid GA/PSO for the concurrent design of cellular manufacturing system

Abstract

In this paper, a hybrid search algorithm using genetic algorithm (GA) and particle swarm optimization (PSO) is implemented for the concurrent design of cellular manufacturing system. Traditionally, cell formation (CF) and group layout (GL) problems were considered sequentially therefore the results may be optimal in one phase but during implementation of the whole cellular manufacturing, it may not be globally optimal. Based on the studies by earlier researchers concurrent approach does indeed lead to better solution quality than the sequential approach by a magnitude of 2% to 20%. Three performance measures are considered to evaluate the proposed method. They are to minimize total inter-cell and intra-cell moves, total cell load variation and total inter cell moves of part families. The performance of the proposed hybrid GA/PSO is evaluated with the test problems available in the literature. The results obtained clearly indicate the better performance of the proposed heuristic.