Abstract

Traffic congestion remains one of the most critical challenges in urban transportation systems, particularly in rapidly growing metropolitan environments. Accurate traffic flow prediction is essential for intelligent transport management and congestion mitigation. This study proposes a hybrid computational intelligence framework integrating Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), and XGBoost for enhanced traffic flow prediction. The model leverages big data analytics and ensemble learning techniques to optimize hyperparameters and improve predictive accuracy. ACO, PSO, and ABC are employed as metaheuristic optimizers to fine-tune the XGBoost learning parameters, thereby minimizing prediction error and enhancing convergence speed. Experimental evaluation demonstrates that the hybrid ACO–PSO–ABC-XGBoost model significantly outperforms conventional machine learning models in terms of Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R² score. The results confirm that integrating swarm intelligence with gradient boosting provides a robust and scalable solution for real-time traffic forecasting.