Inference from Accelerated Life Tests Using the Inverse Power Law Model.

摘要

Parameters of the Power Rule Model, and the shape parameter of the Weibull distribution are estimated using data from censored sample life tests conducted at accelerated environments. Two approaches to the maximum likelihood method are considered; one is based on the observed time to failure and the other is based on the maximum likelihood estimators of the Weibull scale and shape parameters. The behavior of the likelihood function for the latter approach is examined in good detail. The asymptotic normality of these estimators is ascertained by examining the shapes of the relative likelihood functions. Inference about the Weibull scale parameter at use conditions environment is discussed. A numerical example based on accelerated life test data is presented and discussed. (Author)