Geo/Geo/2 Queues and the Poisson Clumping Heuristic

Abstract

In discrete time, customers arrive at random. Each waits until one of two servers is available; each thereafter departs at random. We seek the distribution of maximum line length of idle customers. In the context of an emergency room (for medical treatment), the virtue of one fast doctor over two slow doctors is explored. Via limiting argument to continuous time, we study likewise the M/M/2 queue.