CONICET | Buscador de Institutos y Recursos Humanos

In the last years, the operational research on scheduling problems has been moving away from rigorous optimization approaches into solution strategies being capable of returning practical and fast solutions for large-scale industrial problems. Following this line, this paper proposes a novel MILP-based decomposition procedure for solving scheduling problems arising in flexible manufacturing environments, which generally involve multipurpose units and assembly operations. This solution strategy also incorporates redesign onstraints with the goal of improving the efficiency of the production system, preventing bottlenecks and balancing the equipment utilization. The proposal is validated through the resolution ofseveral instances derived from three real-world case-studies coming from different industrial sectors. The computational results show that the decomposition procedure is capable of generating high quality solutions, sometimes the optimal one, with minimum computational effort for all problem instances considered.