The multiscenario lot size problem with concave costs

The dynamic single-facility single-item lot size problem is addressed. The finite planning horizon is divided into several time periods. Although the total demand is assumed to be a fixed value, the distribution of this demand among the different periods is unknown. Therefore, for each period the demand can be chosen from a discrete set of values. For this reason, all the combinations of the demand vector yield a set of different scenarios. Moreover, we assume that the production/reorder and holding cost vectors can vary from one scenario to another. For each scenario, we consider as the objective function the sum of the production/reorder and the holding costs. The problem consists of determining all the Pareto-optimal or non-dominated production plans with respect to all scenarios. We propose a solution method based on a multiobjective branch and bound approach. Depending on whether shortages are considered or not, different upper bound sets are provided. Computational results on several randomly generated problems are reported.