Optimal incentive-compatible pricing for M/G/1 queues

This paper extends previous research on congested service facilities to generalized service distributions, a significant extension given the limitations of exponential distributions for networked computer job modeling. Building on the framework first presented in Mendelson and Whang (1990), we present fundamental theorems for non-priority M/G/1 queues, nonpreemptive M/G/1 queues, and preemptive-resume M/G/1 queues. For non-priority M/G/1 queues and nonpreemptive M/G/1 queues, these theorems establish optimal assignment rules and incentivecompatible pricing schemes. For preemptive-resume M/G/1 queues, we prove that the total delay cost is less than the total delay cost of nonpreemptive M/G/1 if service times are heterogeneous with decreasing failure rates. This result supports the traditional delay cost to service time ( i i c v ) assignment rule as a good heuristic for preemptive-resume M/G/1 queues. Subject categories: Queues: Priority; Probability: Stochastic model applications