Effect of Capacity Tightness on Performance of MOPSO Algorithm: Case of Multi-item Capacitated Lot-Sizing Problem

This paper investigates the effect of capacity tightness on performance of multi-objective particle swarm optimization (MOPSO) algorithm in solving the multi-item capacitated lot-sizing problem with consideration of setup times and backlogging (MICLSP-SB). The considered problem is formulated as a multi-objective optimization model. The formulated model aims at simultaneously minimizing two objective functions. The first one seeks to minimize the total cost, which the sum of production, setup and backlogging costs. The second one seeks to minimize the total inventory level. Sensitivity analysis is performed on a set of generated problem instances. The capacity tightness factor is defined as the ratio between the required capacity and the total available capacity. Three levels of capacity tightness factor are considered for each problem instances. The metrics, which are used for evaluating the performances of the MOPSO algorithm, are number of Pareto solutions, spacing and computational time. Results of sensitivity analysis show a considerable impact of capacity tightness on performances of MOPSO algorithm in terms of number of Pareto solutions and computational time. This investigation offers to the decision makers a clear insight into the performances of MOPSO algorithm in solving the considered MICLSP-SB depending on problem features especially, the capacity tightness factor.