Bluebike Station Siting Project Summary

The team researched best practices in bikeshare network planning, and collaborated closely with the Transportation Department’s Bikes Team to incorporate subject matter expertise into the analysis. 

We used two primary data sources for our analysis: U.S. Census Bureau survey data and ridership and station status data from Bluebikes. The team also created some of our own metrics from the Bluebikes station data:

•  “Lost trips” - estimates of how many trips didn’t occur because there are no bikes available at the station. A similar “lost returns” metric was created to identify the impact of periods of times with no available docks. By estimating these metrics for each station for each half hour window of the day, we were able to target where empty or full stations were causing the greatest inconvenience to Bluebikes users. For example, a station near a downtown office building being empty in the morning before work hours might not result in many “lost trips” since few morning commute trips start downtown, whereas a station in a residential neighborhood empty at the same time could indicate a severe shortage of available bikes for potential users. We used these metrics to target pockets in the city where existing demand was greater than the current system could meet. The interactive map below shows lost trips and lost returns across the City. 
• Service areas and dock density - these geographic features were generated to show what areas are within walking range of existing Bluebikes stations. We used this to highlight areas with low access to the Bluebikes network.

In order to provide the Transportation Department with specific recommendations for station locations, the team used a suitability modeling approach in GIS. The model takes input data layers and overlays them on top of each other, weighting and synthesizing the different layers into a single score. We could then use these scores to map the best locations for new stations.

In parallel to the suitability modeling approach, the team also used a GIS overlay analysis to get another perspective on the data and identify places with strong potential for new stations. For example, areas with high population density and with low station access could potentially be good candidates for new stations. 

The analysis resulted in a series of recommendations that range from proposed station locations, to more generalized suggested station counts by neighborhood. Currently, the Transportation Department is working closely with the Bluebikes operator Lyft to finalize their expansion plan. Existing street layouts and infrastructure, community engagement, and the Analytics Team’s work will all play a role in determining where stations are sited, and the team is excited to support the Transportation Department further throughout the evolution of this process.