Adaptive Fuzzy Logic Integration for Optimizing Decision Support Systems under Data Uncertainty

Abstract

Decision Support Systems (DSS) become inaccurate when used with imprecise, incomplete, or dynamically changing data. Fuzzy logic techniques based on conventional methodology may be strong at handling vagueness, but are unable to adapt their behavior to different data distributions on their own. This paper introduces an Adaptive Fuzzy Logic Integration Framework that dynamically updates membership functions and rule weights in response to data variation to enhance decision accuracy under uncertainty. The described framework combines Fuzzy Inference Systems (FIS) with learning-based parameter update concepts borrowed from adaptive optimisation. The model was simulated and executed on a hybrid algorithmic platform that included gradient-based parameter tuning and iterative feedback learning. Experimental tests on uncertainty-generated datasets demonstrate that the adaptive model achieves a mean accuracy gain of 21.4% and a 28% improvement in convergence rate compared to non-adaptive fuzzy systems. Moreover, the model ensures stable performance even in the presence of random data perturbations, demonstrating its responsiveness and robustness under uncertainty. The framework provides a self-tuning fuzzy decision model that transforms static inference structures into dynamic, evolving decision engines, establishing a foundation for next-generation smart DSS for real-time optimization.