Optimal path recommendation in dynamic traffic networks using the hybrid Tabu-A* algorithm

Abstract

Recommended routes serve as the cornerstone of intelligent transportation systems, enabling efficient navigation in dynamic traffic environments. Traditional methods model the problem as a route-finding problem on dynamic graphs; however, they often suffer from heuristic inaccuracies and a tendency to become trapped in local optima. To address this challenge, this paper introduces Tabu-A*, a hybrid algorithm that integrates A*'s heuristic cost estimation with Tabu Search's global optimization capabilities. Within this framework, search efficiency is improved while incorporating the best route from each iteration accelerates convergence. Real-world distance and time data enhance adaptability to traffic variations. The algorithm achieves up to a 78.77% reduction in travel time compared to the shortest-path route and improves route duration efficiency by 65.77% over benchmark methods such as A*, Dijkstra, and Bellman-Ford. These results validate the effectiveness of the proposed approach in delivering time-efficient and congestion-aware route recommendations in dynamic environments.