AI-Enabled Digital Twin Framework for Safe and Sustainable Intelligent Transportation

This study proposes an AI-powered Digital Twin (DT) platform designed to support real-time traffic risk prediction, proactive decision-making, and foster sustainable, resilient mobility management in smart cities. The proposed system integrates multi-source data—including static infrastructure, historical traffic records, and real-time IoT and camera feeds—into a unified cyber-physical platform. Leveraging AI models for anomaly detection, risk prediction, and eco-driving optimization, the DT generates dynamic safety heatmaps, predicts congestion patterns, and enables scenario-based decision support. A key feature of the framework is its modular API architecture, allowing seamless integration with existing Advanced Traffic Management Systems (ATMS) and urban data platforms. A pilot deployment on Madison’s Flex Lane corridor demonstrates the system’s ability to process real-time data streams, reconstruct traffic incidents, and generate actionable insights such as collision risk forecasts and adaptive routing strategies. These capabilities provide traffic operators and policymakers with critical tools to optimize signal timing, coordinate emergency responses, and evaluate policy interventions under varying scenarios. By bridging predictive analytics with real-world deployment, this research provides a scalable approach to enhance urban mobility resilience, improve traffic safety, and enable sustainable transportation planning under rising uncertainties.