A survey of research on the simulation startup problem

Abstract

To reduce the effects of initial conditions on the results of a simulation experiment, several authors have proposed startup policies that specify how to minimize the warmup period and identify the trunca tion point beyond which observations are to be re corded and analyzed. Other approaches to this prob- Zem have used results from time-series analysis and queueing theory. This paper surveys research on the simulation startup problem; a companion paper which follows presents a procedure for evaluating startup policies when the mean of the truncated sample is to be used to estimate the steady-state mean of the underlying process.