Photo by EuanFisk on Flickr.
Travelers in the DC region have many modes to choose from. When, and why, do people choose one mode over another? Answering these questions can help transportation planners improve each mode of travel. We compared travel times between Capital Bikeshare and transit to start understanding one aspect of this question.
Transportation professionals have never had all the data and tools they’d like to really study travel patterns and alternatives, but by using trip planners in creative ways, they can gain new insights into people’s behavior. OpenTripPlanner, an open source trip planning system created by OpenPlans, can not only help users navigate their cities, but aid planning as well.
Are riders saving time by using a bike rather than taking transit? Is there a time of day when transit is more competitive than biking because of wait time? We ran a random 1% sample of Capital Bikeshare’s trip history data through OTP to see how biking compared to transit and walking.
This graph compares the actual trip time for each CaBi trip on the vertical axis and OTP’s predicted transit trip time for that same origin and destination on the horizontal axis.
The 45° red line is an approximate indifference line. When a trip takes very long by bike but very short by transit, it shows up above the indifference line; if it takes long by transit but short by bike, the point would fall below the line.
The vast concentration of trips fall in an area below the curve where a bike-share trip is faster than it would take to ride transit. This might logically suggest that CaBi riders use the service when its faster than transit, and not to use CaBi when it would take longer. However, we would need more data from non-CaBi riders and non-CaBi/transit trips to conclude this more definitively.
People taking trips that are significantly longer by bike than by transit are probably taking a leisure ride, or are lost. The triangular white space on the bottom right shows the limits of bike speeds. For example, a bike cannot take a 40 minute transit trip in 5 minutes or less.
Just how fast are CaBi trips?
If you ride CaBi, you’ll notice that after a certain point, you just can’t pedal any faster. CaBi bikes have their gears set for lower speeds.
We also ran the trips through OTP’s bike routing system to find out the most logical travel distance by bike between two stations, then figured out the speed based on the time the rider checked out the bike.
The average trip speed (station-to-station time) is just under 8 mph. The histogram shows the handful of very slow speeds (again, lost or leisure riders) and a nice distribution of fast and slower speeds.
How do you get the fastest speeds? We didn’t run the analysis, but as a former Arlington resident, I suspect some may be a quick trip that starts in Courthouse and ends at the bottom of the hill in Rosslyn.
Bikeshare time doesn’t count walking time
There are some important caveats in the methodology.
The trip data provides a start and end station ID, which we matched to a location using the latest live station feed. The bike share stations in DC are extremely modular and often get moved around from one side of an intersection to another or a block away. This means that there may be slight discrepancies between where trips started in the past and where we project them starting based on today’s feed, but that difference is likely small.
More significantly, the CaBi data only identifies the bike portion of a trip. A person actually walks to a bikeshare station, retrieves the bike, rides to another station, returns the bike and walks to their ultimate destination. When we use these coordinates to plan transit trips, it will return an itinerary for an entire transit trip which includes walking access time.
This is equivalent to having a bikeshare station at your front door and in your office building, then comparing that bike share trip to the time it takes to walk to a Metro station, take Metro and walk to the office.
This simplifying assumption presents a challenge for us that we’d love to get some feedback on. How can we synthesize a set of origin-destination coordinates based on the actual station-to-station bike share trips? Or how can we find a proxy for walk-time that would make this comparison more equal?
What are your ideas?
This analysis doesn’t explain everything about displaced transit trips, but provides an interesting starting point for this type of analysis and how transportation planners can use OTP. We’re looking forward to finding more ways to explore trip data using OTP. If you have ideas to share, please comment here! If you’re interested in doing this kind of analysis yourself, you can get all the scripts and files at this github repository.