USING FUZZY LOGIC IN DECISION-MAKING SYSTEMS

Abstract

In today’s conditions of rapid development of information technologies and increasing complexity of decision- making processes, methods that allow taking into account uncertainty and incompleteness of information are becoming increasingly relevant. One of such methods is the use of fuzzy logic, which provides the possibility of formalization and analysis of complex systems, where classical approaches are ineffective. This article is devoted to the development of methodological principles and practical recommendations for the implementation of decision-making support systems based on fuzzy logic The key advantages of the fuzzy logic approach over traditional methods are investigated, in particular the ability to model uncertainty, operate with qualitative linguistic variables and take into account expert knowledge in the decision- making process. It is argued that fuzzy logic provides a more natural description of real systems with their inherent imprecision and uncertainty, which is especially relevant for DSSs that operate in dynamic environments with incomplete information.Special attention is paid to the methodological aspects of integrating fuzzy logic with other intelligent technologies, including neural networks, genetic algorithms and machine learning methods, which allows creating hybrid systems with increased adaptability and efficiency. Architectural solutions for building DSSs based on fuzzy logic are analyzed with a detailed consideration of the stages of fuzzification, fuzzy inference and defuzzification. The article presents the results of practical implementation of the proposed approaches in various subject areas, including financial management, medical diagnostics, technical systems management and environmental monitoring. A comparative analysis of the effectiveness of DSS based on fuzzy logic with traditional systems is presented according to the criteria of accuracy, speed and interpretation of results.The practical significance of the study lies in the development of specific methods for designing and implementing fuzzy DSS, which can be used by specialists in various industries to improve the efficiency of decision-making processes in conditions of uncertainty and incompleteness of input data