An importance sampling scheme for simulating the degradation and failure of complex systems during finite missions

Authors:
Leonidas C. Kioussis;Douglas R. Miller
Affiliations:
Department of Operations Research, The George Washington University, Washington, D.C.;Department of Operations Research, The George Washington University, Washington, D.C.
Venue:
WSC '83 Proceedings of the 15th conference on Winter Simulation - Volume 2
Year:
1983

Citing 0
Cited 3

A decomposition approach to variance reduction

WSC '85 Proceedings of the 17th conference on Winter simulation
Variance reduction in simulation

WSC '84 Proceedings of the 16th conference on Winter simulation
Simulating markov-reward processes with rare events

ACM Transactions on Modeling and Computer Simulation (TOMACS)

Quantified Score

Hi-index	0.00

Visualization

Abstract

The variance reduction technique of path-splitting is applied to reliability analysis of complex systems. Two approaches are given. In the first, paths are split into a fixed prespecified number of branches. In the second, a sequential procedure is used to estimate the number of branches which will give the maximum variance reduction. The method is illustrated on a simulation of a simplified model of a fault-tolerant computer system. The estimated variance reduction ranged from 36% to 61%.