An efficient reliability system for censoring the data based on the hybrid approach

Abstract

In software reliability modeling, the model parameters are computed from the test data. By removing the uncertainty of the parameter the uncertainty in the total system reliability can also be underestimated. The reliability of manufactured items often involves performing a life test and examining observed times of failure. The lack of existing failure data from a single test makes modeling difficult. This is overcome by combining entropy principle and Kullback Leibler method. In this paper reliability is tested for Type II censored data by combining Kullback Leibler method and Bayesian method. This paper provides aspects of Kullback Leibler information with the introduction of Weibull distribution and Bayesian method for computing risk in Type II censoring scheme. Kullback-Leibler determines the decomposition of entropy in censoring schemes using the Weibull distribution. The method involves estimating the posterior distribution of the Kullback-Leibler information. In our paper we are computing the reliability of the system and analyses the risk involved. The relative risks due to each cause of failure are investigated. The risk of the censoring scheme is analyzed.