ABSTRACT
One method of measuring the reliability of a product is to test a batch of items over an extended period of time and to note the failure times. This process is called life-testing. There are two ways of accelerating life testing. The first method, called compressed-time testing, requires that the product should be used more intensively than usual but without changing the stress levels to which the product is subjected. The second method of accelerated life-testing is called advanced-stress testing and consists of subjecting the product to higher stress levels than would normally apply. Mann et al. discuss the fitting of statistical failure models, accelerated life-testing and system reliability. Gross and Clark are concerned with fitting survival distributions to biomedical life-time data. Data obtained from accelerated life testing is of course much less reliable than that obtained under normal conditions and should be treated with suspicion.
