Far East Journal of Theoretical Statistics
Volume 27, Issue 2, Pages 233 - 246
(March 2009)
|
|
STATISTICAL INFERENCE ON ESTIMATING THE FAILURE INTENSITY OF A WEIBULL PROCESS
Kuan-Fu Shen (Taiwan) and Yu-Jan Shen (Taiwan)
|
Abstract: Predictive modeling in the field of reliability is usually based on the failure intensity function of repairable systems. A Weibull process/non-homogeneous Poisson process is commonly used to analyze the failure behavior of repairable systems. The object of this study is to obtain estimates of the failure intensity of this model at the time of n failures. The resulting MLE estimate is biased and the alternative version along with some approximate estimates is given. A Bayesian statistical approach is presented, which yields posterior distributions of the parameters of the power law which are used to model the trend in the data observed. Simulation studies about approximate estimates of the intensity function are given to illustrate the proposed inferential procedure. |
Keywords and phrases: failure intensity, non-homogeneous Poisson process (NHPP), quasi-Bayes estimation. |
|
Number of Downloads: 253 | Number of Views: 704 |
|