Improved Self-Learning Genetic Algorithm for Solving Flexible Job Shop Scheduling

Abstract

The flexible job shop scheduling problem (FJSP), one of the core problems in the field of generative manufacturing process planning, has become a hotspot and a challenge in manufacturing production research. In this study, an improved self-learning genetic algorithm is proposed. The single mutation approach of the genetic algorithm was improved, while four mutation operators were designed on the basis of process coding and machine coding; their weights were updated and their selection mutation operators were adjusted according to the performance in the iterative process. Combined with the improved population initialization method and the optimized crossover strategy, the local search capability was enhanced, and the convergence speed was accelerated. The effectiveness and feasibility of the algorithm were verified by testing the benchmark arithmetic examples and numerical experiments.