Sudden death testing versus traditional censored life testing. A Monte-Carlo study

This paper considers two competing methods intended to shorten lifetime tests. The first method, due to L.G. Johnson, is known in reliability engineering as "sudden death testing". Its competitor is a widely known time-terminated, right-censored test. Times of tests carried out according to these methods are set equal. Then, methods are compared in terms of variances and biases of lifetime parameter estimators. In addition, median, mode, skewness and kurtosis of estimator distributions are also calculated and compared. All data needed came from a large-scale Monte-Carlo numerical experiment.