An inverse evolutionary algorithm with a multi-operator local search method for flexible job-shop scheduling problem with fixture loading and unloading constraints

Optimizing resource allocation in discrete manufacturing is critical for production efficiency. On-site machining fixtures, as both production tasks and resource constraints, impose non-negligible loading/unloading times, yet such constraints are rarely addressed in flexible job-shop scheduling (FJSP). This paper investigates the FJSP with on-site fixture loading/unloading constraints (FJSP-FLU) to minimize total weighted tardiness (TWT). This paper proposes IGA4FLU, an integrated framework combining: (1) an inverse genetic algorithm with novel operators to enhance exploration and (2) a multi-operator variable neighborhood search to improve exploitation. A fixture-based decoding method is developed to incorporate loading/unloading times, ensuring feasible scheduling solutions. Experiments on 20 benchmark instances and a real-world case validate IGA4FLU’s superiority. Compared to state-of-the-art algorithms, it achieves up to 68.27% TWT reduction and 49.64% makespan improvement, demonstrating its effectiveness in complex manufacturing scenarios.