Edge computing-based mathematical model for latency reduction in urban transportation systems

Managing transportation systems in modern cities is increasingly complex owing to growing urbanization, rising vehicle numbers, and limited road infrastructure. High traffic density during morning peak hours makes intelligent control systems based on real-time data processing essential. Traditional cloud-based architectures suffer from network latency, bandwidth saturation, and delayed decision-making when large data volumes are transmitted to remote servers. This article proposes a mathematical model based on the edge computing paradigm to minimize network latency in cyber-physical transportation systems. In the model, primary data processing is performed at edge nodes located near road intersections, thereby reducing the data volume transmitted to the cloud and increasing real-time decision-making speed. The model was experimentally evaluated on real Los Angeles traffic data (METR-LA, 207 sensors). Results show that edge–cloud integration reduces the mean end-to-end latency from 132.4 ms to 39.9 ms (approximately 69.8%) and cuts bandwidth usage by 80%, confirming that the edge-based architecture contributes substantially to more stable and efficient traffic management during morning peak hours.