Manufacturers are often faced with the problem of how to select the most reliable design among several competing designs in the stage of development.It becomes complicated if products are highly reliable. Under the circumstances, recent work has focused on the study with degradation data by assuming that degradation paths follow Wiener processes or random-effect models. However, it is more appropriate to use gamma processes to model degradation data with monotone-increasing pattern. This article deals with the selection problem for such processes. With a minimum probability of correct decision, optimal test plans can be obtained by minimizing the total cost.
Communications in Statistics-Theory and Methods 43(10-12), pp.2419-2428