Development of a multi-stage fuzzy cognitive map for an uncertainty environment: methods and introduction

Today, hospital failures become a challenging issue regarding their occasional irreparable consequences. Many patients lose their lives each year due to medical failures, with their costs estimated at millions of dollars. Therefore, paying attention to them and correct management of these failures can save the patients&rsquo life. In this regard, various methods have been introduced for evaluating and ranking them. Failure Mode and Effects Analysis (FMEA) in combination with the Gray Relational Analysis (GRA) is among the most effective methods for evaluating these failures. The present study aims to evaluate patient operating failures by considering such failures in an uncertain environment and considering causal relations between these failures. For this purpose, a new method is proposed, which is an extension of the fuzzy cognitive map. This method, called multi-stage fuzzy gray cognitive map, evaluates the operating room failures more accurately by considering the values obtained from FMEA as input, as well as the causal relationship between the failures identified by the medical expert team. The proposed cognitive map was trained by two network training methods including delta rule extended for the multi-stage fuzzy gray cognitive map and Nonlinear Hebbian learning (NHL) algorithm. One of the most important limitations of the present study was the lack of sufficient research on the subject under study and also the difficulty of selecting the appropriate case to validate the proposed model. The results obtained from ranking with the proposed approach and comparing it with previous methods showed more accurate ranking and elimination of duplicates at one level of ranking. These results can be a strong reference for management team decisions in future for better management of hospital failures.