Modeling of failure prediction Bayesian network based on FMECA

To establish practical failure prediction model and facilitate its applications, this paper proposed a failure prediction Bayesian network (FPBN) modeling method based on the failure mode, effects and criticality analysis (FMECA). According to the useful failure relationships embedded in the FMECA unit, the FPBN network construction and the probability parameter computation algorithms were discussed to build the corresponding FPBN unit model. Then, the FPBN unit model of each part in the complex equipment was connected with each other to establish the integrated FPBN model of the whole system. At last, based on the FMECA knowledge of a head up display (HUD), the practical FPBN case of the electronic part was built for the actual failure prediction tasks. The case analysis results show that, the FMECA knowledge embedded FPBN model, which has the advantages of uncertainty representation and quantitative analysis, can perform effectively in the failure prediction of complex equipments.