Investigation of effect due to correlation between components on system reliability
Loading...
Authors
Ayers, David R.
Schwarz, Ira N.
Subjects
Advisors
Woods, W. Max
Date of Issue
1963
Date
Publisher
Monterey, California: U.S. Naval Postgraduate School
Language
en_US
Abstract
System reliability estimates are generally made using a model
which assumes independence between components, and results are often
claimed to be conservative. For a two component serial system bivariate
distributions are developed for three cases: (1) bivariate exponential,
(2) bivariate geometric, and (3) a composite exponential,
geometric bivariate. These distributions are then utilized to investigate
the reliability of a two component serial system when an
estimate of the correlation coefficient is available. The estimate
of the reliability thus obtained is then compared with the corresponding
estimate obtained by use of the model which assumes that the system
reliability is the product of the component reliabilities. The difference between these two estimates is tabulated for values of the
correlation coefficient between -.25 and +.25, for each of the the bivariate distributions. Conditions under which the effect is maximum
are explored and a method of approximating the reliability difference
is suggested. The writers wish to express their appreciation for the assistance
and encouragement given them by Professor Walter Max Woods of the
U. S. Naval Postgraduate School in this investigation.
Type
Thesis
Description
Series/Report No
Department
Organization
Identifiers
NPS Report Number
Sponsors
Funder
Format
Citation
Distribution Statement
Rights
This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.