An effective Markov random fields based estimation of distribution algorithm and scheduling of flexible job shop problem

During the past several years, a large number of studies have been conducted in the area of flexible job shop problems. Intelligent manufacturing planning and scheduling solutions that are based on meta-heuristic methods, such as the simulated annealing and particle swarm optimization, have become common techniques for finding satisfactory solutions within reasonable computational times in real scenarios. However, only a limited number of studies have analyzed the effects of interdependent relationships associated with various decision factors considered for the complex problems. This paper presents a Markov network based estimation of distribution algorithm to address the flexible job shop scheduling problem. The proposal uses a subclass of estimation of distribution algorithms where the effects between decision variables are represented as an undirected graph model. Furthermore, a critical path-based local search method is adopted by the proposed algorithm to achieve better performance. We present an empirical validation for the proposal by applying it to solve various benchmark flexible job shop problems.