Real-time route optimization in smart cities via Bidirectional A* algorithm

Efficient route planning is crucial to intelligent transportation systems, particularly in congested urban environments where the shortest-distance path may often overlook factors such as real-time congestion, which can lead to significant delays during peak hours. This paper presents an end-to-end, graph-based framework for low-latency route computation on urban road networks whose edge travel times change with traffic. Using a publicly available road network for Thessaloniki and historical speed measurements, we construct traffic-updated, time-sliced travel time graphs and support two operational objectives: (i) shortest-distance routing and (ii) fastest-time routing under a given traffic snapshot. As an algorithmic backbone, we employ Bidirectional A* (BiA*), a classical optimal search technique, and effectively deploy it in a traffic-updated setting through careful weight construction, admissible lower bounds for time weights, and a robust preprocessing pipeline that preserves intersection connectivity. Experiments across diverse origin–destination pairs and peak/off-peak periods show that the proposed framework saves time, reaching 82.52% in the peak period when selecting the recommended route. The BiA* also reduces query latency relative to Dijkstra and standard A*. Against additional baselines, Bellman–Ford and improved ACO, across seven scenarios and both objectives, BiA* achieves the lowest runtimes while improving the route length and travel-time of the trip under identical traffic snapshots. An ablation study on route direction shows that search direction can affect performance; nevertheless, BiA* consistently outperforms the mean of forward- and backward-only A*, reducing runtime by 23.6–66.2% for shortest-distance queries and 37.0–87.5% for minimum-travel-time queries, with the largest gains for time-weighted fastest-route searches.