A Review of Dynamic Traffic Flow Prediction Methods for Global Energy-Efficient Route Planning

The rapid growth of urbanization and vehicle ownership has exacerbated global traffic congestion, leading to increased fuel consumption, greenhouse gas emissions, and reduced travel efficiency. While dynamic traffic flow prediction and energy-efficient routing have seen significant progress—leveraging statistical models, machine learning, and deep learning for spatiotemporal analysis, and eco-routing algorithms for energy optimization—these fields remain largely disconnected. This review systematically evaluates traffic prediction methods (2020–2025), benchmarking them using Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), computational latency, and scalability. We then propose an integrated framework that embeds real-time traffic forecasts into path search algorithms (A*, Dijkstra, genetic algorithms) via a multi-objective cost function optimizing distance, time, and energy consumption. Key challenges—data inconsistency, real-time IoT/5G deployment, and multimodal scalability—are discussed alongside future research directions. By unifying prediction and planning, this work provides a roadmap for next-generation intelligent transportation systems, advancing sustainable and efficient urban mobility.