Optimal burn-in time for highly reliable products

Burn-in test has been widely used by many manufactures of electronic products to eliminate latent failures or weak components. However, the traditional burn-in test over a short period of time to collect time-to-failure or go/no go data becomes rather ineffective. In this paper, we use a diffusion process to describe the degradation path of a suitable quality characteristic, highly correlated with the product's lifetime. A decision rule for classifying a unit as a normal or a weak unit is proposed. An economic model is then used to determine the optimal termination time of a burn-in test. Finally, an example of an electronic product is used to illustrate the proposed procedure.