A simple and effective evolutionary algorithm for multiobjective flexible job shop scheduling

This paper addresses the multiobjective flexible job shop scheduling problem (MOFJSP) regarding minimizing the makespan, total workload, and maximum workload. The problem is solved in a Pareto manner, whose goal is to seek for the set of Pareto optimal solutions. We propose a multiobjective evolutionary algorithm, which utilizes effective genetic operators and maintains population diversity carefully. A main feature of the proposed algorithm is its simplicity—it needs only two parameters. Performance of our algorithm is compared with seven state-of-the-art algorithms on fifteen popular benchmark instances. Only our algorithm can find 70% or more non-dominated solutions for every instance.