A simulated annealing algorithm for flexible job-shop scheduling problem

Abstract

This study addresses the flexible job-shop scheduling problem to minimize makespan. In fact, the FJSP mainly presents two difficulties. The first one is to assign each operation to a machine out of a set of capable machines and the second one is to sequence the assigned operations on all machines. Hence, to solve this NP-hard problem, a simulated annealing algorithm is proposed. The meta-heuristic algorithm explores the solution space using a stochastic local search while trying to avoid local optima through accepting probabilistic moves to the worse solutions. The neighborhood search structures of assignment and sequencing are used for generating neighboring solutions to search the solution space. To evaluate the performance of the algorithm, twenty benchmark problems adopted from the literature are used. Consequently, the computational results validate the quality of present approach. © 2009 Asian Network for Scientific Information.