An FMEA Risk Assessment Method Based on Social Networks Considering Expert Clustering and Risk Attitudes

Failure mode and effect analysis (FMEA) method has been widely utilized to solve the problem of risk assessment in all walks of life. An FMEA decision support model considering expert clustering and risk attitude is constructed. First, expert risk assessment information is processed in cloud environment. The clustering behavior of experts is simulated based on trust relationship, opinion similarity, and risk attitude similarity. Second, consensus opinions are formed through opinion evolution, and the group weight determination model is constructed considering the group size and consensus level. Finally, a linear programming model minimizing individual regret is used to solve the risk factor (RF) weight problem. Combined with regret theory and the TODIM method considering finite rationality, the priority of risk is determined. The novel FMEA approach is applied to address reliability management problem of smart bracelets. Sensitivity and comparative analyses demonstrated the effectiveness and superiority of this method and enrich the theoretical research of the FMEA approach.