A MULTI-OBJECTIVE OPTIMIZATION APPROACH FOR ERGONOMIC U-SHAPED MIXED-MODEL ASSEMBLY LINE BALANCING PROBLEM

Abstract

A multi-objective ergonomic U-shaped mixed-model assembly line balancing problem is formulated by integrating operational efficiency and human-centered performance within a unified optimization framework. The model simultaneously minimizes cycle time, maximum ergonomic risk, and ergonomic class imbalance, enabling the joint evaluation of workload intensity and its distribution across stations. Ergonomic risk is quantified using the energy expenditure (EE) method, providing a continuous and decision-dependent assessment of physical workload. Two Pareto-based metaheuristic algorithms, Non-dominated Sorting Genetic Algorithm II (NSGA-II) and Multi-objective Particle Swarm Optimization (MOPSO), are employed under a consistent solution representation and feasibility mechanism. Computational results indicate an average reduction of 10.9% in cycle time for MOPSO relative to NSGA-II, particularly for medium- and large-scale instances. In addition, MOPSO yields comparable ergonomic risk and class imbalance values, while producing a wider and more diverse Pareto front. Lower computational times are also observed, indicating improved solution efficiency. The results suggest that the proposed framework provides an effective and scalable approach for simultaneously addressing productivity and ergonomic performance in complex assembly systems.