A relaxation time approach for reducing initialization bias in simulation

  • Authors:
  • Emily Roth;Alan H. Rutan

  • Affiliations:
  • Department of Economics, Operations Research and Statistics, GTE Laboratories Incorporated;Department of Economics, Operations Research and Statistics, GTE Laboratories Incorporated

  • Venue:
  • ANSS '85 Proceedings of the 18th annual symposium on Simulation
  • Year:
  • 1985

Quantified Score

Hi-index 0.00

Visualization

Abstract

We introduce a truncation algorithm to reduce the effects of initialization bias on Monte-Carlo simulations of stationary Markovian queueing systems. The heuristic is based on the estimated relaxation time of the underlying model and implementation requires no preliminary test runs or parameter selection. Computational results are presented for M/M/1 systems in which performance of the relaxation time method is compared with other truncation rules.