Predicting risk criticalities of hospital sterilization using fuzzy FMEA and artificial neural network

Abstract

The Failure Mode and Effects Analysis (FMEA) is a method widely for managing risks. FMEA uses 3 factors, namely the occurrence, the non-detection and the severity, to calculate the criticality of each risk. This method has proved good effectiveness. However, it was largely criticized due to several weakness. This study aims to develop a model that improves the FMEA method using fuzzy inference system and the artificial neural network. This model is applied in the case of a central sterilization unit of the largest university hospital in Morocco. The proposed model has proved high efficiency: 97.8% is the accuracy rate given by all datasets. Furthermore, the minimal error performance is equal to 14.1%. This study will be generalized to other hospital departments to improve the model’s effectiveness.