Modelling, analysis and simulation for the design of a robotic assembly system

Abstract

This paper describes the modelling, analysis and simulation of a complex assembly system that employs advanced conveyor controls and intricate pallet routing decisions. The modelling and simulation have been carried out using a general-purpose simulation language, SIMAN. A major problem of the initial system design found from simulation results is the congestion caused by using too many pallets in the system. Conveyor congestion causes resource starvation (resources waiting for their pallets to arrive), which in turn requires more workers than necessary to keep up with the production rate. To eliminate the need for exhaustive simulation runs, a simplified analytical model based on a closed-loop queuing network was used to derive the optimal number of pallets circulating in the system and then compared with that based on the simulation model. The results of analyses were used to identify necessary modifications so as to minimize the need for expensive system redesigns and fine tunings. Copyright © 1996 Elsevier Science Ltd.