<p>This study paved the way for developing digital twins of smart and emerging urban mobility systems, using shared mobility services such as ridesharing as a key case study. As cities contend with challenges, such as traffic congestion, environmental sustainability, and transportation equity, shared mobility platforms (e.g., UberPOOL and Lyft Shared) have emerged as promising solutions. Leveraging Chicago’s Transportation Network Companies (TNCs) shared mobility data set, this research uncovers latent patterns in user behavior and trip-sharing dynamics through data mining and exploratory analysis. It distinguishes between trips, where users authorized ride-sharing and those that were actually pooled, revealing key spatial, temporal and behavioral difference. Economic factors also played an important role. For instance, the hourly gap between authorized and successfully pooled trips was narrower on weekends, suggesting more stable matching opportunities, while users who authorized but were not pooled tended to pay less per mile than the general trip population. Building on these insights, this study integrates both supervised and unsupervised machine learning methods to enhance the understanding of ridesharing dynamics. Density-based spatial clustering of applications with noise (DBSCAN) was employed to uncover latent trip groupings, which served as the foundation for developing predictive models that estimate the likelihood of successful ride matches. Multiple classifiers, including Logistic Regression, Random Forest, and XGBoost, were implemented and rigorously evaluated to identify the most effective predictive model. This integrated approach not only provides a comprehensive perspective on ridesharing behavior and trip shareability within current mobility platform, but also builds the foundation for early stage digital twins that can simulate, optimize, and inform decision-making in future smart mobility systems, including autonomous vehicle fleet operations.</p>