A0612
Title: Progressive stress accelerated life testing under unified progressive hybrid censoring
Authors: Aman Prakash - Sardar Vallabhbhai National Institute of Technology, Surat, India (395007) (India) [presenting]
Raj Kamal Maurya - Sardar-Vallabhbhai-National-Institute-of-Technology-Surat (India)
Abstract: The focus is on progressive-stress accelerated life testing (ALT) under a unified progressive hybrid censoring scheme. It is assumed that the lifetimes of test units follow a Lomax distribution, with the scale parameter modeled using the cumulative exposure model and an inverse power law. The model parameters are estimated using both classical and Bayesian methods. The classical approach uses maximum likelihood estimation for point estimates and asymptotic confidence intervals. In contrast, Bayesian estimation employs informative and noninformative priors, with the Metropolis-Hastings algorithm handling complex posterior computations. Extensive simulations evaluate the performance of various progressive-stress ALT, including simple and multiple ramp-stress tests. The above methodology is applied to real-life data, and its applicability is demonstrated. Furthermore, machine learning techniques are introduced to analyze the reliability of the model. Three machine learning methods are employed, such as random forest regression, decision tree regression, and multilayer perceptron regression, to estimate reliability. A comparative analysis of all three methods using performance metrics has been discussed. Furthermore, the machine learning model is compared with the statistical model for reliability estimation.
