Fitting creep-rupture life distribution using accelerated life testing data

This paper comments on using the Larson-Miller parameter to fit the creep-r upture life distribution as a function of temperature and stress. The commo nly used least-squares linear regression method assumes that the creep-rupt ure life follows the lognormal distribution. Most engineering literature do es not discuss the validity of this assumption. In this paper, we outline t he procedure for validating two critical assumptions when the least-squares method is used. The maximum likelihood method is suggested as an alternati ve and more powerful method for fitting creep-rupture life distributions. E xamples are given to demonstrate the use of these two methods using Microso ft Excel and the LIFEREG procedure in SAS. [S0094-9930(00)00504-7].