A Survey of Techniques for Determining Lower Confidence Bounds on Series System Reliability Based on Subsystem Test Data
Abstract
**Please note that the full text is embargoed** ABSTRACT: A problem of considerable interest and for which a
great deal of research has been expended is that of
determining lower confidence bounds on series system
reliability based on subsystem failure data. With
regard to the types of failure data taken on the
subsystems, primary consideration is given to
binomial (pass-fail) data, exponentially distributed
time-to-fail data for both Type I censoring (fixed
test times) and Type II censoring (fixed number of
failures), or any mixture thereof. In this survey we
shall examine optimum solutions whenever they exist,
approximate optimum solutions, and nonoptimum
solutions. Numerical examples are given to
illustrate the theory.