Clone this wiki locally
Popular in Europe and Asia, bike share has landed in the United States with systems popping up in Chicago, Boston, DC, and New York. Soon San Francisco and Seattle will join these cities with systems of their own.
To help bikeshare operators rebalance these systems, we're making an app that predict the number of bikes at a station in next hour or two. The app is live at bikeshare.dssg.io.
This wiki is the central place to learn about the bike share rebalancing problem we're solving, the predictive models and data we're using, our model results, and the web app we built.
- Problem: Bikeshare Rebalancing
- Data: Real-Time and Historical Bike Share System Status
- Exploratory Analysis: Station Statistics and Information
- Methodology: A Series of Statistical Models
- Results: Validating our Models
- Rolling Forward: Future Work
- Webapp: A Tool for Bike Share Rebalancing Dispatchers