Leveraging Digital Twins and Generative AI for Effective Urban Mobility Management

Rapid urbanization and population growth have created significant challenges in urban mobility management, including traffic congestion, inefficient public transportation, and environmental pollution. This paper presents the development and implementation of a Digital Twin (DT) for smart mobility, designed to address these issues. The DT platform integrates a diverse range of historical and real-time data, providing a comprehensive view of urban mobility conditions. Descriptive statistics are employed to identify patterns in parking occupancy and violation frequencies, while Machine Learning and Deep Learning algorithms enhance predictive and generative analytics for forecasting parking needs and simulating scenarios with other mobility aspects. Advanced analytics uncover hidden patterns and behaviors, and visualization tools map data onto the urban layout to facilitate spatial planning and resource allocation. Scenario simulation enables urban planners to assess the impact of different strategies in a virtual environment prior to real-world implementation. The integration of Generative Artificial Intel- ligence (GenAl) models further enhances predictive capabilities and scenario generation. Preliminary results demonstrate the DT platform's potential in improving urban mobility management, particularly in optimizing parking meter placement and enhancing user experience. Although limited data availability affects long-term prediction accuracy, the model exhibits robustness and adaptability for extended forecasting horizons.