In empirical engineering projects on reliability testing, replacement strategy, spares for repair and reliability growth, the Mean Residual Life (MRL) function may be more relevant than the hazard function or failure time distribution function, although they are mathematically equivalent. However, for modeling purposes, MRL and the hazard functions play somewhat different roles. The MRL summarizes the entire residual life distribution of time t, and the hazard function only relates to the risk of immediate failure at time t. When the main concern is the risk of immediate failure, the Proportional Hazards (PH) model has proved to be an extremely useful model. Yet, it may not be the most appropriate model when one is concerned with the remaining lifetime of an individual part at time t as in replacement and repair models. In this paper, a new accelerated life testing model based on the Proportional MRL (PMRL) model in the presence of more than one failure mode is presented and its applicability in empirical reliability engineering is shown. The model can provide accurate reliability estimates for multiple failure modes problems, and is a useful alternative to the Accelerated Failure Time (AFT) and the PH models. In some cases, our model is even better than the PH model.

Keywords