Resumen

EN-NAAOUI, Amine; GALLAB, Maryam  y  KAICER, Mohammed. An intelligent model for improving risk assessment in sterilization units using revised FMEA, fuzzy inference, k-Nearest Neighbors and support vector machine. J. appl. res. technol [online]. 2023, vol.21, n.5, pp.772-786.  Epub 23-Ago-2024. ISSN 2448-6736.  https://doi.org/10.22201/icat.24486736e.2023.21.5.2116.

Risk assessment is an essential decision-making issue in the healthcare sector. Our study aims to improve the process of risk assessment in healthcare organizations by adapting the failure mode and effects analysis (FMEA) to the studied context (revised FMEA), improving criticality calculating using fuzzy logic, and performing tolerance classification with machine learning algorithms. The application area of the model is the sterilization unit of a university-tertiary hospital. The performance of the proposed model is evaluated as follows: we extensively explored the literature to compare fuzzy FMEA with classical FMEA. The results showed that the fuzzy approach is more effective than the classical. Furthermore, some SVM classifiers have been able to achieve 100% accuracy in both training and testing datasets, and the KNN classifier has achieved 97% and 75% of accuracy in training and testing data, respectively. This study will be applied to other hospital departments to generalize our model.

Palabras llave : risk assessment; FMEA; fuzzy inference system; support vector machine; k-nearest neighbor; sterilization unit.

        · texto en Inglés     · Inglés ( pdf )