Title :
Simulating a Weibull posterior using Bayes inference
Author :
Giuntini, Ronald E. ; Giuntini, Michael E.
Author_Institution :
Wyle Labs., Huntsville, AL, USA
Abstract :
A means of deriving a reasonable estimate of reliability is presented for situations where there are no applicable data. The process simulates a Weibull posterior distribution which approaches the true cumulative failure distribution. A Bayesian inference process incorporates less than perfect prediction data (the sample distribution) by means of a heuristic combining function. Through an iterative procedure, each successive posterior becomes a better estimator of the true failure distribution. The process can be employed at any desired system level, and is not costly or labor intensive. It can be implemented with or without the aid of a computer
Keywords :
Bayes methods; iterative methods; reliability theory; simulation; Bayesian inference; Weibull posterior distribution; cumulative failure distribution; heuristic combining function; iterative procedure; reliability estimation; simulation; Bayesian methods; Costs; Laboratories; Life estimation; Maintenance; Modeling; Monte Carlo methods; NASA; Product development; Shape;
Conference_Titel :
Reliability and Maintainability Symposium, 1993. Proceedings., Annual
Conference_Location :
Atlanta, GA
Print_ISBN :
0-7803-0943-X
DOI :
10.1109/RAMS.1993.296877