MODELING FUZZY SYSTEMS FOR DECISION-MAKING UNDER UNCERTAINTY

The article discusses methods for modeling fuzzy systems to support decision-making under uncertainty. The relevance of using fuzzy logic is substantiated, allowing for the integration of expert knowledge and taking into account linguistically described parameters, which significantly increases the adaptability and accuracy of systems when processing incomplete or approximate data. The theoretical foundations of fuzzy logic are presented, including the concepts of fuzzy sets, membership functions and aggregation operators, and a comparative analysis of the Mamdani and Takagi-Sugeno models is carried out. Experimental studies have demonstrated that optimization of the parameters of membership functions (e.g., logistic, Gaussian, and triangular) and the selection of appropriate defuzzification methods help to reduce the forecast error and increase the stability of the system. The article highlights the prospects for further development of hybrid models that combine fuzzy logic with machine learning and optimization methods, which opens up new opportunities for creating universal decision support systems under conditions of high uncertainty.