Statistical Analysis from the Generalized Inverse Lindley Distribution with Adaptive Type-II Progressively Hybrid Censoring Scheme

Abstract

The key assumption in accelerated life testing is that the mathematical model concerning the lifetime of the item and the stress is known or can be assumed. In several situations, such life-stress relationships are not known and cannot be assumed, i.e. accelerated life testing information cannot be extrapolated to use situation. So, in such cases, a partially accelerated life test is a more appropriate testing method to be executed for which tested objects are subjected to both normal and accelerated circumstances. Due to continual improvement in manufacturing design, it is more difficult to obtain information about the lifetime of products or materials with high reliability at the time of testing under normal conditions. An approach to accelerate failures is the step-stress partially accelerated life test which increases the load applied to the goods in a particular discrete sequence. In this study, the maximum likelihood estimators of inverse the generalized inverse Lindley distribution parameters and the acceleration factor are investigated in a step-stress partially accelerated life test model utilizing two various types of progressively hybrid censoring systems. Furthermore, the performance of the model parameter estimators with the two progressive hybrid censoring schemes is analyzed and compared in terms of biases and mean squared errors using a Monte Carlo simulation approach.