Application of Bayesian networks for product quality management in a multistage manufacturing process

Nowadays, requirements of clients and customers for the quality of product are more and more tightened and complicated. The quality assurance of manufactured product is a key to success in the context of global and competitive economy. Many different parts of final product are made from raw material by multistage manufacturing processes in different places. The risk is that the final manufactured product does not fully meet the requirements. Thus, the paper proposes a method based on Bayesian networks that allows to model impact factors in a multistage machining process on product quality. The root cause analysis can be implemented by using the Bayesian network model. As a result, product quality predicted earlier at design stage can help product designer adjust the product designed and manufacturing processes in order to obtain a robust design with promised quality.