A comparison of degradation and failure-time analysis methods for estimating a time-to-failure distribution

Degradation analysis can be used to assess reliability when few or even no failures are expected in a life test. In this paper, we use a simple but useful degradation model to compare degradation analysis and traditional failure-time analysis in terms of asymptotic efficiency. The comparisons consider a range of practical testing situations and provide insight into the trade-offs between these two methods of estimating the quantiles of the time-to-failure distribution. We investigate the effect that the number of inspections, the amount of measurement error, and the quantile of interest have on the asymptotic variances of the quantile estimators. Although measurement error can induce some loss of precision in degradation analysis, our comparisons show that, except in extreme cases, degradation analysis provides more precision than traditional failure-time analysis.