Arranging queues in series: A simulation experiment

For given arrival process and given service-time distributions, the object is to determine the order of infinite-capacity single-server queues in series that minimizes the long-run average sojourn time per customer. The authors gain additional insight into this queueing design problem, and congestion in non-Markov open queueing networks more generally, by performing simulations for the case of two queues. For this design problem, they conclude that the key issue is variability: The order tends to matter more when the service-time distributions have significantly different variability, and less otherwise. Arranging the queues in order of increasing service-time variability, using the squared coefficient of variation as a partial characterization of variability, seems to be an effective simple design heuristic. Parametric-decomposition approximations seem to provide relatively good quantitative estimates of how much the order matters.