Variable neighborhood search for the capacitated lot sizing problem with remanufacturing and overtime

This study proposes a mixed integer linear programming (MILP) model for the single-item multi-period capacitated lot-sizing problem with remanufacturing. The tackled problem is capacitated in terms of production and storage. The production capacity is overcome by allowing for overtime in some periods. Moreover, the customer demand is assumed to be dynamic and deterministic. Demand is satisfied by either producing new items or remanufacturing returned items. The objective is to obtain an optimal plan over a finite planning horizon. The structure of the problem results in strongly NP-hard problem. Consequently, we develop a variable neighborhood search (VNS) metaheuristic algorithm for solving this problem. We demonstrate the robustness of the proposed VNS approach by examining different problem instances with different sizes. The obtained results are compared with the output of CPLEX solver in GAMS. The results obtained from the proposed metaheuristic outperform the solver output in terms of solutions quality and computational time.