Modelling accelerated life testing based on mean residual life

Abstract

Accelerated life testing (ALT) is a widely used approach for reliability demonstration and prediction. Extensive research on ALT models has been focused on Accelerated Failure Time (AFT) models, Proportional Hazards (PH) models and some extensions of these two models. Mean residual life function provides a more descriptive measure of the aging process than the hazard function; however, it has not been used in modelling ALT. In this paper, we propose an ALT model based on a Proportional Mean Residual Life (PMRL) model and demonstrate its applicability in the reliability field. The model utilizes accelerated conditions data to estimate the reliability measures under normal operating conditions and provides a viable alternative to the accelerated failure time (AFT) model and the proportional hazards (PH) model. Some results concerning aging properties for the PMRL model are also studied.