Summ. & Conclusions - This paper develops a Raves model for step-stres
s accelerated life testing. The failure times at each stress level are
exponentially distributed, but strict adherence to a time-transformat
ion function is not required. Rather, prior information is used to def
ine indirectly a multivariate prior distribution for the failure rates
at the various stress levels. Our prior distribution preserves the na
tural ordering of the failure rates in both the prior gr posterior est
imates. Methods are developed for Bayes point estimates as well as for
making probability statements for use-stress life parameters. The app
roach is illustrated with an example.