AI Safety

NetMob26 Dataset: 7.2M bus transactions unlock Niterói mobility patterns

GPS telemetry plus 7.2M ticket records reveal public transit in unprecedented detail.

Deep Dive

The NetMob Data Challenge has released NetMob26, a high-resolution multi-source public transportation dataset from Niterói, Brazil. Covering March 2026, the dataset fuses GPS telemetry from buses with approximately 7.2 million ticketing transactions, plus auxiliary transit data (routes, stops, weather) and urban infrastructure/socio-demographic information. This combination offers a detailed view of both transit supply and passenger demand. The data was preprocessed, cleaned, and anonymized to remove operational inconsistencies and protect passenger privacy. Access is limited to challenge participants who sign an NDA and accept terms.

Researchers Felipe Domingos and colleagues describe the collection pipeline, dataset organization, and observed mobility patterns in a 39-page arXiv paper. The dataset supports studies on public transportation efficiency, demand forecasting, accessibility analysis, service reliability, and the influence of external factors like weather on urban mobility. By providing both supply-side (bus telemetry) and demand-side (ticketing) data at high resolution, NetMob26 fills a critical gap in public transit research, enabling more accurate models and policy recommendations for cities.

Key Points
  • Combines GPS telemetry from buses and 7.2M ticket transactions from March 2026 in Niterói.
  • Includes transit routes, stops, weather data, and socio-demographic info for multi-factor analysis.
  • Access restricted to NetMob Challenge participants under NDA; data anonymized and cleaned for privacy.

Why It Matters

Enables data-driven improvements in public transit efficiency, demand forecasting, and urban accessibility worldwide.