Response surface analysis of stochastic network performance

  • Authors:
  • T. G. Bailey;K. Bauer;A. B. Marsh

  • Affiliations:
  • -;-;-

  • Venue:
  • WSC '89 Proceedings of the 21st conference on Winter simulation
  • Year:
  • 1989

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes a response surface approach to the stochastic network improvement problem. The network addressed in this study is acyclic, simple, and directed; and, is characterized by single commodity flow from multiple sources to multiple sinks. The network is stochastic in that the reliability of network components is described by a binary distribution of operative or failed states, where Psi is the ith component's probability of survival. A new class of control variables is introduced to reduce the variance of the maximal flow estimators.