Abstract:
|
The production of semiconductor devices for the automotive industry is characterized by high reliability requirements, such that the proper functioning of these devices is ensured over the whole specified lifetime. Therefore, manufacturers let their products undergo extensive testing procedures that simulate the tough requirements their products have to withstand. Such tests typically are highly accelerated, in order to test the behavior of the products over the whole lifetime. In case of drift of electrical parameters, manufacturers then need to find appropriate guard bands for their production processes, such that the proper functioning of their devices over the whole specified lifetime is ensured. In this study, we present statistical models for optimal guard bands. The models consider longitudinal measurements of continuous features, based on censored data from stress tests. Guard bands are derived from multivariate distributions where the dependence structure is described by different copulas. Based on extensive numerical testing of our guard banding procedures, we are able to provide insights on the benefits and drawbacks of our models under several production conditions.
|