Abstract
In this paper a new generalization of the Wilcoxon test is proposed for the comparative reliability evaluation of alternative engineering designs. Using Monte Carlo simulations the performance of the new test is assessed against the results provided by four other two-sample tests for censored reliability data, i.e., the Gehan-Wilcoxon, Peto-Prentice, Logrank and modified Kolmogorov-Smirnov tests. The results of the simulations suggest that the Wilcoxon alternatives based on weight functions that are reliability function estimates, such as the Peto-Prentice and the Bohoris-Wilcoxon, are the most preferable for routine use.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have