Predicting the performance of rescheduling strategies for parallel machinesystems

In dynamic, stochastic manufacturing systems, production planners and manuf acturing engineers can benefit from understanding how rescheduling strategi es affect system performance. This knowledge will help these experts design and operate better manufacturing planning and control systems. This paper presents new analytical models that can predict the performance of reschedu ling strategies and quantify the trade-offs between different performance m easures. In the parallel machine systems under consideration, jobs of diffe rent types arrive dynamically, and setups occur when production changes fro m one job type to another. Three rescheduling strategies are studied: perio dic, hybrid, and event-driven based on the queue size. The scheduling algor ithm groups jobs of the same type in batches to eliminate unnecessary setup s. The analytical models require less computational effort than simulation models, and experimental results show that they accurately estimate importa nt performance measures like average flow time, machine utilization, and se tup frequency.