A new test for testing that a life distribution is exponential against the alternative that it is harmonic new better (worse) than used in expectation upper tail HNBUET (HNWUET), but not exponential is presented based on the highly popular 'Kernel methods' of curve fitting. This new procedure is competitive with old one in the sense of Pitman's asymptotic relative efficiency, easy to compute and does not depend on the choice of either the band width or kernel. It also enjoys good power.

This paper deals with the stochastic analysis of a three-states semi-Markov reliability model. Using both the maximum likelihood and Bayes procedures, the parameters included in this model are estimated. Next, assuming that the lifetime and repair time are generalized exponential random variables, the reliability function of this system is obtained. Then, the distribution of the first passage time of this system is discussed. Finally, some of the obtained results are compared with those available in the literature.

An economic life test sampling plan for products with exponential lifetime distribution is developed. To reduce test time, a test plan with curtailed Type II censoring is considered. A cost model is constructed which involves three cost components; test cost, accept cost, and reject cost. Determination of optimal plan minimizing the expected average cost per lot is discussed with a constraint related to consumer's risk. Some numerical examples are provided.

The Bayes estimators of the parameters included in the complementary Weibull reliability model are obtained. In the process of deriving Bayes estimators, the scale and shape parameters of the complementary Weibull distribution are considered to be independent random variables having prior exponential distributions. The maximum likelihood estimators of the desired parameters are derived. Further, the least square estimators are obtained in closed forms. Simulation study is made using Monte Carlo method to make a comparison among the obtained estimators. The comparison is made by computing the root mean squared errors associated to each point estimation. Based on the numerical study, the Bayes procedure seems better than the maximum likelihood and least square procedures in the sense of having smaller root mean squared errors.

In this paper, we propose a nonparametric test procedure for the multivariate, grouped and right censored data for two sample problem. For the construction of the test statistic, we use the linear rank statistics for each component and apply the permutation principle for obtaining the null distribution. For the large sample case, the asymptotic distribution is derived under the null hypothesis with the additional assumption that two censoring distributions are also equal. Finally, we illustrate our procedure with an example and discuss some concluding remarks. In appendices, we derive the expression of the covariance matrix and prove the asymptotic distribution.