Dynamic lot-sizing for remanufacturing systems with multi-level structured products and components commonality

This study addresses dynamic lot-sizing for remanufacturing systems in which end-of-use/life products are disassembled into their components on a disassembly workstation, each recoverable part is reprocessed on one of parallel reprocessing workstations and the reprocessed and newly purchased ones, if required, are reassembled into remanufactured products on a reassembly workstation. The problem is to determine disassembly, reprocessing, reassembly lot-sizes as well as the reprocessing workstation to reprocess each recoverable part while satisfying dynamic remanufactured product demands and workstation capacities in each period of a planning horizon. As an extension of the previous studies, multi-level structured products with components commonality are considered for the objective of minimizing the sum of setup, operation and inventory costs. A mixed integer programming model is developed, and due to the problem complexity, two types of heuristics are proposed that obtain an initial solution using a greedytype heuristic and improve it by two types of bi-directional moves with various priority rules for forward and backward moves, i.e. individual-moves and cluster-moves. Computational results show that the cluster-move based approach that considers components commonality explicitly outperforms the individual-move based one significantly and gave fast near optimal solutions for small sized test instances. Finally, a sensitivity analysis on different multi-level product structures were done and the results are reported.