For some specific degradation processes a model involving independent increments is appropriate. In some other cases stochastic process models such as the Gamma and Wiener processes are used. However, there are many applications in which both Wiener and Gamma processes may not be adequate. The inverse Gaussian process is another natural choice for degradation data which provides a monotone degradation path and many articles are available regarding this. However, the inverse Gaussian process and its applications are not discussed much so far. This paper discusses a class of inverse Gaussian process models for degradation data and associated maximum likelihood inferences. A simple graphical method to assess the adequacy of different stochastic process models is also provided. (27 refs.)
展开▼
