This study presents a performance analysis of a fuzzy rule-based system for real-time traffic prediction,
comparing it with traditional and machine learning models. The fuzzy logic approach utilizes a set
of "if-then" rules to incorporate uncertainties and imprecise data, enhancing prediction accuracy. We
evaluated various models, including traditional regression, decision trees, support vector machines,
recurrent neural networks, and convolutional neural networks, using metrics such as accuracy, mean
absolute error (MAE), and computation time. The experimental results indicate that the fuzzy rulebased
system achieved an accuracy of 82.4% and an MAE of 2.1, outperforming traditional models and
demonstrating competitive performance relative to advanced machine learning techniques. Notably, the
recurrent neural network and convolutional neural network provided the highest accuracy but required
longer computation times. This research underscores the potential of fuzzy logic in traffic prediction,
offering a robust alternative to conventional methods while addressing the complexities of dynamic
traffic conditions.
