Abstracts - faqs.org

Abstracts

Business, general

Search abstracts:
Abstracts » Business, general

A variant of the conditional expectation variance reduction technique and its application to the simulation of GI/G/1 queues

Article Abstract:

The conditional expectation (CE) variance reduction technique is well-known but unpopular due to its model-dependency and the lack of a general method to find a random variable Y for all computed variables of Y. A modified version of the CE technique, the Partial Conditional Expectation (PCE), provides the capacity to obtain a consistent estimator for computing the waiting time in queue for a single-server queueing system in simulations of a GI/G/1 queue. An improved version of the estimator can be derived by further consideration of the system. The two estimators preform better than the classical estimator and variance reduction obtained from the two estimators is particularly significant with low traffic.

Author: Minh, Do Le
Publisher: Institute for Operations Research and the Management Sciences
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1989

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


Simulating GI-G-K queues in heavy traffic

Article Abstract:

Waiting times in heavy traffic queues are analyzed for length of time, queue length, and expected service backlog. The method used to measure these factors is called an error estimator because it estimates the error involved in the use of Kingman's heavy traffic approximation. The error estimator method is superior to the regenerative error method because it retains the advantages of the latter while substantially lowering variances when heavy traffic is involved. The regenerative method is more versatile in some situations, however.

Author: Minh, Do Le
Publisher: Institute for Operations Research and the Management Sciences
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1987
Methods, Work measurement

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


Importance sampling for stochastic simulations

Article Abstract:

Importance sampling is an effective technique for increasing the efficiency of Monte Carlo algorithms for the numerical analysis of integrals. The original random mechanism in the simulation can be replaced in order to solve problems arising from the simulation of stochastic systems for application to GI/G/1 queueing problems, including: discrete time Markov chains; continuous-time Markov chains; and generalized semi-Markov processes.

Author: Glynn, Peter W., Iglehart, Donald L.
Publisher: Institute for Operations Research and the Management Sciences
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1989
Stochastic analysis, Samples (Merchandising), Samples (Products)

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


Subjects list: Research, Analysis, Management research, Computer simulation, Analysis of variance, Queuing theory
Similar abstracts:
  • Abstracts: Experimental Evaluation of Variance Reduction Techniques for Queueing Simulation Using Generalized Concomitant Variables
  • Abstracts: Multinational corporate restructuring and international competitiveness. Employee work attitudes and management practice in the U.S. and Japan: evidence form a large comparative survey
  • Abstracts: Japanese makers of computer displays may face higher anti-dumping duties. U.S. considers easing export restraints on technological gear to the East bloc
This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.
Some parts © 2025 Advameg, Inc.