Variance reductions in simulation of a state–dependent queueing system using total hazard. (4th August 2014)

• Record Type: Journal Article
• Authors: Jun, Chi–Hyuck; Ross, Sheldon M.
• Is Part Of: International journal of continuing engineering education and lifelong learning. Volume 4:Number 1/2(1994)
• Page Start: 69
• Page End: 79
• Publication Date: 2014-08-04
• Subjects: queueing systems -- simulation -- total hazards -- control variables -- variance reduction -- state–dependent queueing -- Poisson process; Engineering -- Study and teaching -- Periodicals; 620.00715
• Languages: English
• ISSNs: 1560-4624

Abstract

A queueing system is considered where customers arrive according to a Poisson process. Each customer possesses a random workload which is processed by a single server using a state–dependent rate. Total hazards are derived and utilised as control variates for reducing the variance of the raw simulation estimator of the expected number of customers during a busy period. Simulation experiments are done for exponential and k–Erlang workload distributions. Performance of controlled estimators is measured in terms of the mean squared error, which is estimated by the bootstrap method. Simulation results show that hazard–controlled estimators outperform the raw simulation estimator and other controlled estimators based on workloads and inter–arrival times.