From space to automotive: early detection of burnin rejects on SiC products using a new statistical screening approach

Abstract

Quality and reliability are more and more important in automotive industry as the number of components per car increases and will increase dramatically in the next years with the electrical and autonomous vehicles. This is especially true when new technologies are going to be deployed to address power and other challenges, especially the Silicon Carbide (SiC). To support high quality standards, we have developed a new statistical multivariate method called Good Average Testing (GAT), which is an efficient tool for screening outliers, i.e. reliability issues. It is already in use for space industry and we have adapted it to the context of automotive. We propose to show is efficiency on SiC products.