Using state-dependent processing rates to emulate SPT queue discipline in an FCPS queueing network

Abstract

Using the ‘shortest processing time’ queue discipline results in long queues being quickly reduced as the short jobs are quickly pushed through the system. We approximate this effect under the ‘first-come, first-served’ queue discipline by having the service rate increase with the queue length. That is, instead of picking the shortest job, the server instead picks the first job, but then processes it at a higher rate. This new load-dependent queue is compared with the original queue both individually and within open queueing networks.

Notes

Lawrence W. Robinson is an Associate Professor of Operations Management at Cornell University's Johnson Graduate School of Management, where he has worked since receiving his Ph.D. from the Graduate School of Business at the University of Chicago in 1987. His teaching interests include quality management and production and operations management. His research focuses on developing practical policies for operating in an uncertain environment, and ranges from inventory management, to booking limits for discount-fare airline passengers, to scheduling doctors' appointments. He has published in a number of journals, including Operations Research, IIE Transactions, and the European Journal of Operational Research.

Kevin B. Hendricks received his Ph.D. in Operations Management from the Johnson Graduate School of Management at Cornell University in 1990. He is an Assistant Professor of Operations and Information Technology in the School of Business at the College of William and Mary in Williamsburg, Virginia. His current areas of interest include manufacturing planning and control systems, shop floor control, stochastic models of manufacturing systems, queueing networks, and empirical research on the impact of manufacturing strategy decisions such as capacity expansion, product development, and total quality management on firm performance. His research has previously been published in Operations Research and Management Science.