Sensitivity of Output Performance Measures to Input Distributions in Queueing Simulation Modeling

With modern simulation packages, the modeler can choose almost any standard statistical distribution for generating input random variables. The question arises as to how sensitive are output performance measures to the particular type of distributions selected for modeling the input. We investigate the sensitivity of two output performance measures (average queue wait and the 95 percentile value of the queue wait distribution) for a G/G/1 queue, simulated using GPSS/H.

• Type: Conference
• Year: 1997
• Where: WSC
• Authors: Donald Gross, Man Juttijudata
• Keywords: Average Queue Wait, Modeling And Simulation, Output Performance Measures, Queue Wait, WSC 1997