Bicing: Understanding Human Behavior through Shared Bicycle Programs

Bike-sharing systems generate detailed records of when and where bicycles are collected and returned. We used these records to characterize and predict patterns of urban mobility.

The analysis covered six weeks of activity at 373 stations in Barcelona's Bicing system. We grouped stations with similar usage and examined the relationship between those patterns and location, neighborhood, elevation, and time of day. Bayesian-network models then used those signals to predict station usage.

Different parts of the city exhibited distinct daily patterns. These patterns allow planners to estimate demand and anticipate where bicycles or empty docking spaces will be needed.

The records describe one bike-sharing system over a short period, not all travel in Barcelona. Weather, seasonal demand, service changes, and trips made by other forms of transport were outside the dataset, so the station patterns should not be read as a complete model of urban mobility.

A nearly full Bicing station, a station kiosk, and a bicycle locked into a docking point. A Bicing station kiosk. A bicycle secured at a Bicing station.

Barcelona's Bicing infrastructure.

Map showing the locations of 373 Bicing stations across Barcelona.

The 373 stations included in the study.

Visualization grouping Bicing stations by similar usage patterns. Chart comparing station elevation with the average number of available bicycles.

Stations were grouped by shared behavior and compared with features of their surroundings.

Publications

Conference paper
DOI
Cite
Formatted citation

Jon Edward Froehlich, Joachim Neumann, Nuria Oliver (2009). Sensing and Predicting the Pulse of the City Through Shared Bicycling. Proceedings of the 21st International Joint Conference on Artificial Intelligence (IJCAI 2009). https://doi.org/10.5555/1661445.1661673

BibTeX
@inproceedings{froehlich2009bicycling,
  author = {Jon Edward Froehlich and Joachim Neumann and Nuria Oliver},
  title = {Sensing and Predicting the Pulse of the City Through Shared Bicycling},
  booktitle = {Proceedings of the 21st International Joint Conference on Artificial Intelligence (IJCAI 2009)},
  year = 2009,
  doi = {10.5555/1661445.1661673},
  cites = 498
}

Sensing and Predicting the Pulse of the City Through Shared Bicycling

Jon Edward Froehlich, Joachim Neumann, Nuria Oliver
Proceedings of the 21st International Joint Conference on Artificial Intelligence (IJCAI 2009) · 2009
Conference paper
External link
Cite
Formatted citation

Jon Froehlich, Joachim Neumann, Nuria Oliver (2008). Measuring the Pulse of the City Through Shared Bicycle Programs. Proceedings of UrbanSense08, 16-20. https://joachimneumann.github.io/publications/UrbanSense08.pdf

BibTeX
@inproceedings{froehlich2008bicycle,
  author = {Jon Froehlich and Joachim Neumann and Nuria Oliver},
  title = {Measuring the Pulse of the City Through Shared Bicycle Programs},
  booktitle = {Proceedings of UrbanSense08},
  pages = {16--20},
  year = 2008,
  url = {https://joachimneumann.github.io/publications/UrbanSense08.pdf},
  cites = 135
}

Measuring the Pulse of the City Through Shared Bicycle Programs

Jon Froehlich, Joachim Neumann, Nuria Oliver
Proceedings of UrbanSense08 · 2008

Press