An Improved Decomposition-Based Multiobjective Evolutionary Algorithm for Human-Centric Flexible Job Shop Scheduling Problem

Overview & Synopsis

In modern manufacturing, integrating human factors into scheduling is challenging due to the inherent variability among workers. Traditional scheduling methods often overlook worker fatigue, leading to suboptimal performance. The main contribution of this study is the development of an improved decomposition based multiobjective evolutionary algorithm (IMOEA/D) to solve the Double Flexible Job Shop Scheduling Problem with Worker Fatigue constraints (DFJSP-WF). Unlike existing studies, the originality of our approach lies in integrating worker fatigue into the scheduling process, aiming to simultaneously optimize makespan and production quality by minimizing errors associated with fatigueinduced human factors. To validate its effectiveness, we compare IMOEA/D with state of the art multiobjective evolutionary algorithms, including NSGA-II and MOEA/D. The results demonstrate its superiority in achieving better trade offs between efficiency and production quality, with improved solution quality and convergence. Furthermore, a comparative analysis with the closely related work, which applied NSGA-II to the classical FJSP with fatigue considerations, confirms the superiority of IMOEA/D in generating higher quality Pareto fronts, as evidenced by improved hypervolume values.