What’s driving transit ridership?

This project demonstrates a speculative transit demand study where I use spatial analytics to help understand the characteristics and trends of MBTA Redline stops in the cities of Cambridge and Somerville. These demand drivers, while simplified, can be quite revealing. In the case of Kendall Square, there is a disproportionate amount of workers opposed to residents when correlating variables – while Porter Square has the opposite problem. Both mismatches can correlate to less forecasted demand for transit when compared to other stations in the area. This example shows how spatially distributing data can guide analysis to help foster better TOD zones or other changes to areas surrounding transit in the effort to encourage public transit use and spur local economic development. This was also one of my earlier projects as a student at the GSD and doesn’t taken into account any statistical regressions… yet.

I started by spatially distributing all residents and employees around each station.

The total sum of both residents and jobs within a 1000m walkshed of each MBTA Redline stop in Cambridge and Somerville.

First Takeaway

This analysis first revealed that the Kendall / MIT and Porter Square stations seem to be below the trend line compared to other observed stations. What could be driving this? Is it the number of residents or employees? Is there reason for concern? Harvard Square is a major tourism hub and students living off campus often take the redline to commute to school, so this might explain Harvard’s higher than usual ridership.

I then decided to distribute the density of jobs first…

This map shoes the density of jobs within each 1000m walkshed. What Kendall Square lacks in residents, it makes up with jobs. This map shoes the density of jobs within each 1000m walkshed. What Kendall Square lacks in residents, it makes up with jobs.

… followed by residents (per sqmi) within a 1000m walkshed of each station.

This map shoes the density of Residents in each 1000m walkshed. Kendall Square has noticeably less residents.

Conclusion

This analysis shows that the Kendall Square station has disproportionate job density, opposed to residential density. Given the relatively high job density, Kendall Square doesn’t perform exceptionally better than the other stations included in the study. This might be explained by Kendall Square industries attracting employees from outside the serviceable area — ie. from the suburbs. Large corporations promising local economic development often fall short of this promise due to a large chunk of their employees commuting in from elsewhere, rather than living nearby. This could be a sign that there isn’t enough housing in Kendall Square, or that the ratio of office space to residential space needs to be adjusted.

Turning over to Porter Square, we find that both the residential and job density in the area isn’t keeping up with the ridership of similar neighboring stations. Porter Square is more of a concern in this regard – since both residents and employees aren’t using transit as one might expect. This could be attributed to an accessibility or design issue in the public realm that deters riders, and should be investigated further.

These are quick analyses that can be done to understand basic drivers of public transportation use and performance. While suggested action cannot be made just yet, this exercise can be useful in understanding where there are service inefficiencies.