Author

Thuan H. Tran

Date of Award

3-1998

Document Type

Thesis

Degree Name

Master of Science in Operations Research

Department

Department of Operational Sciences

First Advisor

Edward A. Pohl, PhD

Second Advisor

W. Paul Murdock, PhD

Abstract

This research explored the applicability of Bayesian methods to the problem of determining the reliability of a series system, a parallel system, and a bridge system. The time to failure of each component in each system is assumed to be exponentially distributed. A basic exponential case and a basic binomial case are also evaluated. In general, using simulated data and real data when possible, the Bayesian methods produced confidence intervals that were much tighter than the classical inference methods, thus allowing a decision maker to have higher confidence in making a decision. System level data and aggregated component level data produced a much tighter confidence interval than component level data aggregated to system.

AFIT Designator

AFIT-GOR-ENS-98M-25

DTIC Accession Number

ADA342127

Share

COinS