Strategic arrivals into queueing networks

Queueing networks are typically analyzed assuming that the arrival process is exogenous, and unaffected by admission control, scheduling policies, etc. In many situations arriving users are strategic, and do time their arrivals taking delay and other metrics into account. This paper builds on, and extends the framework developed to a network setting. We first consider just a single population of users arriving into two queues in parallel (they can join either queue). The queues start serving at different times. We characterize the arrival process into both queues and the Price of Anarchy with strategic arrivals. We then extend this when there are multiple populations, each with different cost metrics. The whole analysis is done in the fluid limit.