The system under consideration is a field replaceable unit referred to as a circuit-pack used in various types of electronic switching equipment. The objective was to analyze field failure data on these circuit-packs in order to obtain estimates of the component reliabilities, which would then be used to predict the reliability of other configurations of the same types of components, i.e. new circuit-packs. In some instances the cause of failure was isolated only to some subset of the components on the pack. An approach to finding maximum likelihood estimates of component life distribution parameters from this type of masked system life data is described. In particular, it is assumed that component life is Weibull-distributed, and an iterative technique for maximizing the complex likelihood expression is presented.<<ETX>>
[1]
Wayne Nelson,et al.
Applied life data analysis
,
1983
.
[2]
N. E. Collins,et al.
Simulated annealing - an annotated bibliography
,
1988
.
[3]
Frank M. Guess,et al.
An iterative approach for estimating component reliability from masked system life data
,
1989
.
[4]
S. M. Alexander,et al.
System reliability prediction based on historical data
,
1990
.
[5]
Frank M. Guess,et al.
Estimating system and component reliabilities under partial information on cause of failure
,
1991
.
[6]
John S. Usher.
A computer program for component life estimation using multiply censored system life data
,
1992
.