A gradient approach for smartly allocating computing budget for discrete event simulation

Simulation plays a vital role in analyzing many discrete event systems. Usually, using simulation to solve such problems can be both expensive and time consuming. We present an effective approach to smartly allocate computing budget for discrete-event simulation. This approach can smartly determine the best simulation lengths for all simulation experiments and significantly reduce the total computation cost for obtaining the same confidence level. Numerical testing shows that our approach can obtain the same simulation quality with one-tenth the simulation effort.