• We're a fun, friendly, and diverse group of planners, placemakers, built environment shapers, students, and other folks who found their people here. Create your FREE Cyburbia ID, and join us today! Register through your Reddit, Facebook, Google, Twitter, or Microsoft account, or use your email address.

Transit Ridership forecasting based on proximity to bus stops and frequency

Doberman

Cyburbian
Messages
180
Points
7
We are trying to get more granular with estimating our ridership for future service plan proposals. Without getting all engineerish, it would seem like there is probably a correlation between ridership, proximity to stops, and frequency. I realize that proximity wouldn't be as the crow flies and you would really need a walkshed analysis that took into account sidewalk networks and topography etc. TCQSM correlates existing ridership with proximity to stops and it also uses frequency as a level of service metric.

It would seem for instance, that you could take a traffic analysis zone and estimate ridership based on frequency and the percentage of the total number of jobs within the proximity of the stop. For example, if your bus was coming every 10 minutes you would forecast maybe 25% of the total number of pops and jobs within 50 meters of your stop as your total ridership.

I realize there are other variables, and a more intense mode of transit such as LRT may have a bigger buffer than fixed-route.

Is there any research or literature akin to what I'm trying to find out?




.quality of service.png

correlation.png
 

glutton

Cyburbian
Messages
416
Points
11
Super interesting question! Sorry I have no answers, but I'd love to learn more about how to forecast transit travel demand. Following!
 
Top