A comparison on two techniques for collecting statistics in simulation

Abstract

The purpose of the paper is to compare two techniques of collecting statistics: The method of batch means and the regenerative process method. The compairsons are in terms of accuracy and precision of sample performance measurements and exexution time.Two different queuing systems of M/M/1 and M/M/4 with traffic intensity factors of 0.1, 0.5, and 0.9 are designed for experimentation. The results indicate that the regenerative method in general shows better performance than the method of batch means.