Simulation based optimization using PSO in manufacturing flow problems: A case study

Abstract

This paper presents the use of simulation based optimization in addressing manufacturing flow problems at a heavy equipments manufacturer. Optimizing the buffer allocation in an assembly line and optimizing the worker assignment at workstations are two independent problems addressed, with the objective to maximize throughput rate. The simulation models of the system, built using an in-house tool based on SLX, is interfaced with a custom designed meta-heuristic based on Particle Swarm Optimization (PSO). Two versions of the PSO have been developed: one with integer decision variables (for buffer space allocation) and another with binary variables (for worker assignment). The performance of the proposed simulation based optimization scheme is illustrated using case studies.