Big data shines light on bike and pedestrian trips

By Chris McCahill

New applications in big data could soon let us understand precisely how people move around by bike and on foot, for all types of trips, almost anywhere in the country. SSTI has worked with several providers to better understand the available trip data and its useful applications. We recently tested preliminary pedestrian data, provided by StreetLight Data, with promising results.

Over the past twenty years, GPS location data has offered valuable new insight about people’s travel patterns. Household travel surveys began incorporating GPS devices in the early 2000s to augment or replace trip diaries. More recently, smartphones let researchers and agencies employ their own apps, such as San Francisco County’s CycleTracks, or rely on data from sources like Strava Metro.

But one great advantage of the newest data is that they’re collected more passively, via apps that run constantly in the background. Cellular location data from providers like Airsage have a similar advantage, but their spatial resolution is much lower. Plenty of GPS location data exist—think of Google and Facebook—but most of them are carefully guarded. StreetLight has bridged that gap, harnessing large amounts of data from many apps and turning them into useable trip metrics, while maintaining strict anonymity standards.

For a recent project, sponsored by TransitCenter, Barr Foundation, and Planet Bike, we looked at first- and last-mile connections to light rail transit in Sacramento using preliminary pedestrian trip data from StreetLight. In one case, we looked at trips to and from Zinfandel station (shown below).

Zinfandel station

Figure 1: Zinfandel station

Some key findings in this example are:

  • Most walking trips (63 percent in total) come from areas just north and east of the station, where there are large shopping centers.
  • 19 percent come from residential areas to the south and west
  • Only 5 percent come from residential areas to the northeast (more use Cordova Town Center station, according to the data)

Since we were mainly interested in opportunities to improve first- and last-mile connections, the most compelling finding was that 15 percent of trips begin or end in a large cluster of office buildings southeast of the station across the Lincoln Highway (US-50). This is somewhat surprising given the abundant parking that serves those buildings, the sparse pedestrian network, and the single access point across the freeway at the Zinfandel Drive interchange (pictured below). An improved crossing could go a long way to improving pedestrian experience in this case.

Figure 2: zinfandel interchange

Figure 2: Zinfandel Drive interchange

There’s also potential for more fine-grained analyses in the future—identifying popular routes or common mid-block crossing points, for example.

Laura Schewel, founder and CEO of StreetLight Data, explains: “After two years of extensive R&D and several pilot projects, we’re very close to formally launching bike and pedestrian Metrics on our StreetLight InSight® platform. Currently, they’re available as a custom order for clients interested in getting the metrics early.” She expects beta metrics to be available via StreetLight InSight by Fall 2017.

Sign up for our newsletter for updates and related reports on this work.

Chris McCahill is a Associate Researcher at SSTI.