Optimal Burn-in Test for Bivariate Degradation Model

Burn-in test is very important in manufacturing process which helps to detect weak items before selling products. Traditional burn-in test which was based on collecting time to failure data is not efficient in short time. When there is at least one quality characteristic (QC) related to lifetime of the products, collecting degradation data and using them in inference can be helpful. New products, due to their complexity, have more than one QC. these QCs may be dependent and so it is very important to find the joint distribution of them. In this paper, we assume that the product has two QCs and their degradation can be governed by inverse Gaussian process. The optimal burn-in test using cost model is obtained and finally a simulation study is conducted.