A Study on Multi-objective Flexible Job Shop Scheduling Problem Using a Non-dominated Sorting Genetic Algorithm

With the fast development of manufacturing digitalization intelligent manufacturing scheduling has become a hotpot which attracts the attention of manufacturers. Among scheduling problems, flexible job shop scheduling problem (FJSP) is a NP hard problem full of difficulty and significance. Moreover, the multi-objective optimization has aroused great interest of company managers. In this paper, an improved non-dominated sorting genetic algorithm (NSGA-II) is proposed to solve the multi-objective FJSP. Firstly, a variable neighborhood structure is utilized as a local search (LS) algorithm to enhance the performance of NSGA-II. Then, crossover and mutation operation are modified to improve the effectiveness of the proposed algorithm. Besides, different datasets are adopted to test the performance of the algorithm. The result shows that the proposed algorithm outperforms NSGA-II in searching for optimal solutions. Furthermore, when deal with benchmark scheduling dataset, the proposed algorithm has a better or an equivalent performance comparing with other intelligent algorithms, such as particle swarm optimization algorithm and a tabu search algorithm(PSO + TS), hybrid tabu search algorithm (HTSA), Pareto-based discrete artificial bee colony (P-DABC). The study provides a meaningful attempt to solve the FJSP in industry production.