Enhanced GO methodology to support failure mode, effects and criticality analysis

Abstract

Failure mode, effects and criticality analysis (FMECA) is a prerequisite and significant task for reliability analysis of a safety critical complex system; meanwhile, it is complicated, error-prone and time-consuming due to the gap between a dedicated process for system functional design and fault propagation analysis. Moreover, traditional empiricism analysis usually omits potential fault propagations, and the Risk Priority Number cannot display the risk of failure modes that attribute to the entire system effectively and precisely. In this article, GO methodology, a wide-spread intuitive reliability and safety model technique in practical engineering systems, is enhanced to support FMECA in a qualitative and quantitative way. Function operators in the GO model are extended to support the function hierarchy, failure modes definition, parameters, failure computation and the connectivity with other components. Based on the information above, the failure propagation algorithm is presented, utilizing the relationship signified by signal flows in the GO model. Next, depending on the extended GO model and automatic FMEA synthesis algorithm, all of the existent and potential failure effects are obtained automatically without artificial experience. With Bayesian Network, the reliability of the system and the risk assessment of each failure mode are obtained simultaneously. Finally, a case study of a mobile platform of robot is introduced to verify and validates the applicability of the proposed method in reliability analysis domains. The proposed method is compared with conventional FMECA approach. It is shown that the proposed method has a better performance in qualitative FMEA analysis and quantitative analysis in risk assessment and sensitivity analysis.