Research on Multi-Objective Hybrid Flow Shop Scheduling Problem With Dual Resource Constraints Using Improved Memetic Algorithm

The classical hybrid flow shop scheduling problem (HFSP) only treats machines as the only resource constraint, ignoring the dominant role of workers in production and manufacturing. Considering the dual flexibility of machine and worker, this paper studies the multi-objective hybrid flow shop scheduling problem with dual resource constraints (DHFSP). Firstly, according to the characteristics of DHFSP and various constraints, the model is built to minimize the maximum completion time (makespan), total tardiness time and workload balance of worker. Then, an improved multi-objective memetic algorithm (IMOMA) is proposed to solve the DHFSP, which mainly includes the improvement of initial population, crossover, mutation and local search. In addition, Taguchi method is used to set parameters. Finally, through numerical experiments, IMOMA is compared with NSGA-II, MODE and MOMVO algorithms. The experimental results show that IMOMA can solve the multi-objective hybrid flow shop scheduling problem with dual resource constraints effectively. In terms of convergence, diversity and dominance of non-dominated solutions, IMOMA is significantly superior to other algorithms, but the distribution uniformity of non-dominated solutions of the four algorithms are not significantly different.