Optimizing Processing Rates for Flexible Manufacturing Systems

Abstract

This paper introduces the generic concept of processing rates as decision variables in Flexible Manufacturing Systems (FMS's). The objective is to determine the minimum cost processing rates given the FMS throughput target, the work-in-process level, part routes, transporter delays, and the variable capacity cost function for each machine. A nonlinear Mean Value Analysis queueing network optimization methodology is developed to control bottlenecks and queue lengths as the processing rates are varied. This methodology further provides the average and marginal unit production costs along with necessary and sufficient feasibility conditions for the FMS throughput targets. Industrial sample data is then used to illustrate the solution of the optimal tool speed problem in a metal-cutting FMS. Considerable cost savings are demonstrated using the proposed methodology in contrast with the conventional one-machine optimization models. Several economic insights regarding the issues of capacity allocation for FMS's, and a generalization of the square root capacity allocation rule for closed networks of queues, are also presented.