Modelling and Optimization of a Flexible Manufacturing System

Abstract

A flexible Manufacturing System (FMS) is simulated using SIMSCRIPT with five workstations. Jobs are assumed to enter and leave an input/output station. Three different types of jobs arrive at the input/output station with interarrivai times that are independent exponential random variables at a certain probability. Each job has a routing system before it leaves the manufacturing system. Service time for each task at a particular workstation is an independent exponential random variable. An Automated Guided Vehicle (AQV) is used to transport the jobs to the workstations. The AGV processes requests by jobs in a First In First Out (FIFO) basis. If the job transported by the AGV to a work station is found to be busy, the job will be placed in a single FIFO queue at that workstation. The FMS facility was simulated for 1000 hours and statistics were gathered for total delay for the jobs in queue, time average number of jobs in queue, average proportion of time that machines were idle, working, or blocked. Analysis were performed to optimize the number of AGVs required for transportation, job arrival rate in anticipation of the demand, and number of machines needed for uniform machine utilization. The results of the analysis are presented in the paper.