Steady state simulation of queueing processes : a survey of problems and solutions

For years computer-based stochastic simulation has been a commonly used tool in the performance evaluation of various systems. Unfortunately results of some studies can be given little credibility, since they are presented neglecting their random nature. On the other hand, a variety of techniques for collection and statistical analysis of simulation data have been proposed that vary in efficiency and sometimes required a deeper knowledge of the statistical methods involved. Practitioners still await a possibly general automatic technique that could be applied in stochastic simulation by users having little knowledge or interest in the statistical analysis of output data. In this report we discuss the main factors that can affect the accuracy of stochastic simulations designed to give insight into the steady state behaviour of queueing processes. The most promising methods for achieving suitably accurate simulation results are also surveyed