Optimum simple ramp test for the log-logistic distribution with censoring

Abstract

This paper presents an optimum simple ramp accelerated life test with two different linearly increasing stresses for the log-logistic distribution under type I censoring. The log-logistic life distribution has been found appropriate for high-reliability components. The inverse power law and a cumulative exposure model are assumed. The higher of the two stress rates is specified and the low stress rate and proportion of test units allocated to the low stress mode are found by minimizing the asymptotic variance of the maximum likelihood estimates of the quantile of the cumulative distribution function at design stress. For some selected values of the parameters, the effect of the incorrect pre-estimates of the design parameters has been studied in terms of the percentage of the variance increase. Confidence intervals involving design parameters have also been obtained.