Improved lower bounds for the proportional lot sizing and scheduling problem

Where standard MLP-solvers fail to compute optimum objective function values for certain MLP-model formulations, lower bounds may be used as a point of reference for evaluating heuristics. In this paper, we compute lower bounds for the multi-level proportional lot sizing and scheduling problem with multiple machines (PLSP-MM). Four approaches are compared: Solving LP-relaxations of two different model formulations, solving a relaxed MLP-model formulation optimally, and solving a Lagrangean relaxation.