Bayesian analysis for system reliability inferences

Due to declining defense budgets, operational test and evaluation programs for military systems have been down-one significant result is less system operational test data. This leads to lower classical confidence inferences on critical measures of performance such as system reliability. This paper e...

Full description

Saved in:
Bibliographic Details
Published inAnnual Reliability and Maintainability. Symposium. 1999 Proceedings (Cat. No.99CH36283) pp. 151 - 157
Main Authors Tran, T.H., Murdock, W.P., Pohl, E.A.
Format Conference Proceeding
LanguageEnglish
Published IEEE 1999
Subjects
Bayesian methods
Bridges
Equations
Exponential distribution
Force measurement
Independent component analysis
Probability
Reliability
System testing
Weapons
Online AccessGet full text

Cover

More Information
Summary:Due to declining defense budgets, operational test and evaluation programs for military systems have been down-one significant result is less system operational test data. This leads to lower classical confidence inferences on critical measures of performance such as system reliability. This paper explores the use of Bayesian methods to compensate for this decrease in test data by integrating developmental test and evaluation data into the operational test and evaluation phase. Specifically, the reliability of a series system, a parallel system and a classic bridge system are examined over a range of mission times. Components in this study are assumed to have exponential lifetimes. Using simulated data for the three systems, the Bayesian methods produce tighter probability intervals for the system reliability than the confidence intervals found through classical inference methods.
ISBN:0780351436
9780780351431
DOI:10.1109/RAMS.1999.744111