Hybrid Artificial Intelligence and Fuzzy Set Theory for Intelligent Decision-Making in Uncertain and Dynamic Systems

Abstract

Modern decision-making systems increasingly operate under conditions characterized by incomplete information, linguistic ambiguity, noisy observations, conflicting criteria, and continuously changing environments. Conventional artificial intelligence models provide strong predictive and optimization capabilities but often exhibit limited interpretability and inadequate representation of imprecise knowledge. Fuzzy set theory, in contrast, offers an effective mathematical framework for representing vagueness and gradual membership but may require substantial expert knowledge and can face difficulties in adapting to rapidly evolving data patterns. This paper develops a hybrid artificial intelligence and fuzzy set theory framework for intelligent decision-making in uncertain and dynamic systems. The proposed approach integrates machine learning-based pattern extraction, fuzzy knowledge representation, adaptive rule generation, uncertainty modelling, multi-criteria evaluation, and dynamic decision updating. The framework is designed to combine the predictive capability of AI with the interpretability and uncertainty-handling capability of fuzzy reasoning. Its applicability is considered across complex decision environments involving heterogeneous data, changing system states, conflicting objectives, and human-oriented decision criteria. The proposed framework provides a unified basis for developing adaptive, transparent, and robust intelligent decision-support systems.