Multi-objective metaheuristic approach for balancing and scheduling human-robot collaborative assembly lines with cognitive and ergonomic considerations

Human-robot collaboration has become a significant approach in assembly lines to enhance efficiency, adaptability, and flexibility. Following the human-centric vision of Industry 5.0, balancing production efficiency with the well-being of human operators is crucial for achieving humane and sustainable assembly. This study addresses the assembly line balancing problem with human-robot collaboration from a human-centric perspective, comprehensively incorporating three objectives: reducing cycle time, balancing cognitive load, and minimising ergonomic risks. An enhanced mathematical model is proposed and validated using data from an automobile assembly line. Given the complexity of the problem, a multi-objective adaptive simulated annealing algorithm with three variants is developed, incorporating a novel adaptive neighbourhood search strategy and a restart mechanism to enhance solution quality and diversity. Computational experiments on case studies based on automobile assembly lines and problems of varying scales demonstrated that the proposed algorithm achieves superior convergence and solution diversity compared with representative state-of-the-art multi-objective metaheuristics, particularly for medium- and large-scale problems. The results indicate that the proposed model and algorithm provide effective decision support for balancing and scheduling human–robot collaborative assembly lines, enhancing the welfare of human operators while maintaining assembly efficiency.