Neighborhood Search Enhanced Nsga-Iii for Multi-Objective Assembly Line Balance Problem Considering Operator Skill Levels and Carbon Footprint

Carbon footprints have been increasingly considered by consumers, making profitability and sustainability the dual mission of a manufacturer. Therefore, this paper considers a multi-objective optimization problem in production lines, focusing on balancing line efficiency, minimizing carbon emissions, and maximizing profits. Additionally, operator skill levels are incorporated and regarded as decision variables. Then, an enhanced version of the Non-Dominated Sorting Genetic Algorithm III, augmented with a neighborhood search strategy (NSGA-III-LS), is proposed to tackle the complexities associated with the proposed model. In particular, the chromosome decoding process is transferred to the one-dimensional variable-size binning problem, which can quickly solve the cycle time of the workstation and the corresponding relationship between the process and the workstation. Comparative experiments demonstrate the effectiveness of the proposed NSGA-III-LS. Also, the performance of the proposed algorithm is validated against other multi-objective optimization approaches.