SynthCity Case Study – Genoa, Italy

🏙️ SynthCity Case Study – Genoa, Italy 📜 Preferred Citation If you use this dataset, please cite the IEEE article below: D. Russo, F. R. D. Torrepadula, L. L. L. Starace, S. D. Martino and N. Mazzocca,“A Framework for Generating Synthetic Urban Mobility Datasets With Customizable Anomalous Scenarios,”IEEE Open Journal of Intelligent Transportation Systems, 2025.DOI: 10.1109/OJITS.2025.3626948 @article{D. Russo, F. Rocco Di Torrepadula, L. Libero Lucio Starace, S. Di Martino and N. Mazzocca, "A Framework for Generating Synthetic Urban Mobility Datasets With Customizable Anomalous Scenarios," in IEEE Open Journal of Intelligent Transportation Systems, vol. 6, pp. 1439-1458, 2025, doi: 10.1109/OJITS.2025.3626948} SynthCity is a multimodal synthetic mobility dataset generated using the Eclipse SUMO traffic simulator. In this case study, we consider open urban data from Genoa, Italy.It enables the analysis, modeling, and evaluation of anomalous urban mobility scenarios, such as strikes, road closures, and sudden demand surges, under realistic city-scale conditions. Built on a modular pipeline for simulation-based synthetic data generation, SynthCity integrates: 🗺️ OpenStreetMap road network 🚍 GTFS data from AMT Genova 🧮 Origin–Destination (OD) matrices from the City of Genoa The dataset includes both nominal and anomalous conditions, providing fine-grained mobility traces for: 🚗 Private transport (individual trajectories, travel times, routes, delays) 🚌 Public transport (stop-level events, passenger counts, delays, service disruptions) 🌐 Network-level dynamics (flow, density, and speed for every edge) 🧪 Anomalous Scenarios Metro Strike — full suspension of subway service and demand redistribution across bus lines. Special Event — localized demand increase around a stadium zone. Road Closure — complete blockage of a key arterial road with dynamic rerouting and congestion buildup. 📊 Dataset Highlights Category Entries Description Public Transport 5,607,953 (normal) / 937,233 (anomalous) Stop-level operations, delays, boarding & alighting Private Vehicles 2,581,172 (normal) / 381,381 (anomalous) Individual trajectories, duration, route length, waiting time Network Edges 58,685,598 (normal) / 8,190,843 (anomalous) Traffic density, mean speed, occupancy, flow Each dataset component is provided as a CSV file. 🧩 Use Cases AI-based mobility prediction and demand forecasting Urban anomaly detection and scenario analysis ITS optimization and resilience modeling Simulation-driven benchmarking for traffic management algorithms