Abstract
For devices with long life, a new and advanced type of stresses test called progressive stress partially accelerated life test (PSPALT) has been adopted to gain quickly information about the device's lifetime distribution. This article presents a PSPALT model when the inverse Weibull life distribution is considered. The progressive stress is assumed to be directly proportional to time. The statistical properties of the maximum likelihood (ML) estimators of the model parameters such as existence, uniqueness and invariance are studied. The mean squared errors of the ML estimators are calculated to evaluate their performances through a Monte Carlo simulation study.