A Bayes study of inverse Gaussian based strength models with accelerating stress.

The survival probability of a composite material subject to external stress is governed by the condition that the strength of the material must exceed the applied stress. There are several models proposed in the literature for studying such stress-strength relationships, especially with reference to modeling failures in carbon fiber composites. A strong contender in this group is the inverse Gaussian based cumulative damage model. The present paper outlines the Bayes analysis of the four models of this family, two for each of additive and multiplicative stress functions. Further a comparative study, based on a real data set, has been carried out to propose the best model among them. [ABSTRACT FROM AUTHOR]