Optimal burn-in policy for highly reliable products using inverse Gaussian degradation process

Burn-in test is a manufacturing procedure implemented to identify and eliminate units with infant mortality before they are shipped to the customers. The traditional burn-in test, collecting event data over a short period of time, is rather inefficient. This problem can be solved if there is a suitable quality characteristic (QC) whose degradation over time can be related to the lifetime of the product. Optimal burn-in policies have been discussed in the literature assuming that the underlying degradation path follows a Wiener process or a gamma process. However, the degradation paths of many products may be more appropriately modeled by an inverse Gaussian process which exhibits a monotone increasing pattern. Here, motivated by the numerous merits of the inverse Gaussian process, we first propose a mixed inverse Gaussian process to describe the degradation paths of the products. Next, we present a decision rule for classifying a unit as typical or weak. A cost model is used to determine the optimal burn-in duration and the optimal cut-off level. A simulation study is carried out to illustrate the proposed procedure.