The objective of many steady-state simulations is to
study the behavior of a nonterminating system with a peak load
of infinite duration. Due to the complexity of the system, the
initial conditions of the system are often atypical that often
requires the simulators to start the system with the empty and
idle conditions. Consequently, deletion of some initial
observations is required to reduce the initialization bias
induced by atypical initial conditions.
This paper studies the application of Schriber's
truncation rule to the complex queueing systems (specifically,
the two-machine and three-machine tandem queueing system) and
the effects of parameter selection (i.e. parameters batch size
and time between observations) on performance measures. Based
on the previous studies of Schriber's rule on the one-machine
system, parameters batch count and tolerance are held
constant.
Mean-squared error and half length are used as measures
of accuracy and interval precision in comparing the results.
The results of both systems show that time between
observations and batch size are significant parameters, and
the recommendations for the two-machine system can be
generalized for the three-machine system. Increasing the
number of machines in the system from two to three requires a
careful reduction in the value of time between observations.
Besides, multiple replications should be used to minimize the
extreme results in determining the steady-state mean number of
entities and the truncation point. / Graduation date: 1992
Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/36568 |
Date | 12 December 1991 |
Creators | Saleh, Budiman |
Contributors | Randhawa, Sabah |
Source Sets | Oregon State University |
Language | en_US |
Detected Language | English |
Type | Thesis/Dissertation |
Page generated in 0.0017 seconds