Generalized step stress accelerated life model [LED case study]

In this paper we develop a general accelerated life model for step stress testing, and present a general likelihood function formulation of step stress models. This model is easy to extend the reliability analysis to multiple-stress and profiled testing data, and reduces to Nelson's cumulative damage model, Gouno (2001)'s model, Xiong and Milliken (2002)'s model as special cases. Algorithms for fitting and testing such models are described and illustrated. The proposed model is applicable on any life distribution as long as the stress level only alters the scale parameter of the distribution. The practical use of the proposed statistical inference is demonstrated by a case study.