Sustainable Food Production: An Intelligent Fault Diagnosis Framework for Analyzing the Risk of Critical Processes

Fault diagnosis and prognosis methods are the most useful tools for risk reliability analysis in food processing systems. Proactive techniques such as failure mode effect (FMEA) important detecting all probable failures facilitating process. However, significant uncertainties exist classical-FMEA when it comes to ranking priority numbers (RPNs) of modes. Such may have an impact on sector’s operational safety maintenance decisions. To address these issues, this research provides a unique FMEA framework within edible oil purification facility that is based certain well-known intelligent models. Fuzzy inference systems (FIS), adaptive neuro-fuzzy (ANFIS), support vector machine (SVM) models among those used. The findings comparison proposed with classical model revealed strategies were more effective RPNs Based performance criteria, was discovered SVM algorithm classifies modes accurately fewer errors., e.g., RMSE = 7.30 MAPE 13.19 other techniques. Hence, sensitivity performed put forward suitable actions upgrade lines.