New Statistic to Increase Correctness in Simulation Factor Screening Using Frequency Domain Method

摘要: Simulation models often include a large number of input factors, many them may be unimportant to the output; justifying use factor screening experiments eliminate factors from consideration in later stages analysis. With challenge is designing so that total runs and consequently required time cost decrease while achieving satisfactory detection rate. This article employs frequency domain method (FDM) which applicable discrete-event simulation propose new statistic defined as ratio estimated signal spectrum maximum noise spectrum. The proposed not only has FDM advantages compared classic approaches but also helps reduce error associated with distinguishing important effects ones. Furthermore, an alternative existing statistics, it shown does deteriorate power ...