Improving Efficiency of the Automated Transportation System in a Large-scale Job-shop Factory using Optimization by Simulation Method

Abstract

We developed a simulation-optimization integration system, for the purpose of improving efficiency of the automated transportation system in a large-scale job-shop factory. In such a factory, many semi-finished goods are delivered to and from shops by automatically controlled vehicles. Those vehicles are required to move efficiently since the traffic jam of vehicles may lower the production efficiency of the shops. In order to implement the simulation-optimization integration system, we used NETLOGO, a multi-agent simulator, and αPSO, an optimization algorithm, which are appropriate for simulating and optimizing such a stochastic complex system as an automated transportation network. We conducted some feasibility studies and found two cases where the efficiencies of the transportation improve: we optimized distance-velocity relation between vehicles, which could increase the flow of the vehicles by 22.1%; we also optimized green period of the signal at the crossing of the transportation, which could reduce the travel time of the vehicles by 12.8%.