Reactive Operations Scheduling for Flexible Manufacturing Systems

Abstract

In this paper we present a reactive operations scheduling methodology for online control of manufacturing systems. We present description of the Texas Instruments electronic circuit board assembly unit at Johnson City, TN, which was used as a model for the development of the reactive operations scheduling scheme. Based on the manufacturing system model, we present the operations scheduling problem as a combinatorial optimization problem. We present a simulated annealing based predictive scheduling algorithm for feed-forward control of the manufacturing system. For reactive operations scheduling, the closed loop configuration consists of the simulated annealing based predictive scheduler and the manufacturing system model in the feed forward path, and a reactive re-scheduler in the feedback path. Deviations of the observed sequenced operation timings from the scheduled timings, quantified by a cost functional, are used to trigger the re-scheduling loop. The capabilities of the scheme are demonstrated using a discrete event simulation of the Texas Instruments manufacturing unit.