Bayesian network for reliability prediction in functional design stage

Due to the lack of reliability data at a product's very early design stage, the reliability prediction in this phase is often problematic. In order to tackle this problem, one needs to utilize auxiliary information such as the reliability information from similar products and domain expertise. In this article, we propose a Bayesian network approach to incorporate data from functional analysis and parent products. We apply our method on a case study to demonstrate our proposed method. In the case study, we identify the functions with low reliability and their impact on other functions in the network, thus design changes can be suggested for system reliability improvement.