Progressively censored reliability sampling plans based on mean product lifetime (Q2188753)

The paper puts forward a design of reliability sampling plans with a Weibull lifetime model based on progressive Type-II censoring in the presence of binomial removals. The Bayes estimator of the mean product lifetime is used for quality assessment. This is done using the Metropolis-under-Gibbs approach under the assumption of two types of loss, i.e. MSE and LINEX. A cost function to determine the Bayes risk and the optimal sampling plan is presented, which includes the sampling, testing time, acceptation and rejection costs. Simulation studies and the application on a real-world data set are performed in the experimental part of the paper.