Organizational Risk Assessment using Adaptive Neuro-Fuzzy Inference System
Javad J. Jassbi, Sohrab Khanmohammadi · European Society for Fuzzy Logic and Technology Conference · 2009
In this paper a fuzzy model based on Adaptive Neuro-Fuzzy Inference System (ANFIS) is introduced for calculating the level of risk in managerial problems. In this model, affecting factors on the level of risk are considered as inputs and the level of risk as the output. Using the fuzzy model, the risky condition changes smoothly in a fuzzy environment as it is the case in the real world; while in classic models, we may have some stepwise changes in the state of the system caused by an infinity small deviations in input parameters. The main advantage of the introduced model is that for continuous values of input factors, the counters of risk surface represent a more realistic behavior for different systems. The model is designed for a system of three inputs (probability, impact and ability to react) and one output (risky situation). The ability of using historical data as well as experts' knowledge and flexibility of adaptation to unusual risky situations are some benefits of the introduced model. This model which is originally used in strategic management system to analyze the external environment and the level of threats can also be used in contingency management for incidents (CMI) or as a tool for Comprehensive Emergency Management Program (CEMP). The strategic Risk of Roche is considered as the bench mark for comparing the difference between fuzzy and classic systems.