Design of Single‐Testing‐Condition Reliability Experiments

This chapter discusses how to use experimental design techniques to improve product reliability. It deals with both product life tests and accelerated life tests under a single testing condition. The most commonly used censoring scheme is Type I censoring or time censoring. Another censoring scheme is Type II censoring or count censoring. Planning life tests with lifetime distributions other than exponential distribution is much harder because, due to censoring, the closed-form solution of confidence interval of the distribution parameter of interest is often elusive. In accelerated life tests, elevated stress conditions are applied on test units to accelerate their failure processes, thus shortening lifetimes. The Arrhenius acceleration model concerns the thermal stress and its effects on material properties. The Peck model is similar to the Eyring model in the sense that both of them are the combination of the Arrhenius function of thermal stress and one or two power functions of other stresses.