Multicampus Parking Study
I work as an institutional researcher at a local college, where I both respond to research requests and act as the data steward for the college. The college is currently updating the facilities master plan and the architecture firm requested a baseline of our parking needs across three campuses. In response to the request, I created a parking demand model that estimated parking demand across all three campuses from 6:15AM to 11pM with an average measurement interval of 17 minutes.
The college is heavily based on commuter students and transit service is not entirely reliable. In addition, the standard techniques for estimating parking demand based on headcount, FTE, paid parking utilization, or parking counts were not available in this situation. Instead, the assumption was that students will need parking when their first class starts and will remain there until the end of their last class unless they need to switch to a different campus*. The study identifies the cumulative need for parking on the student level. Interestingly, this student level of detail would allow for deeper analysis into parking patterns based on different student groups, such as by enrolled program or participation in adult basic ed or the high school program.
To come to these results, I pulled unique students with demographics for the quarter from the student database and then joined to the student class table, where students have a record for each section of a class they are enrolled in. The minimum start time and maximum end time were taken from the results and the campus locations were derived from the building codes associated with each section by linking further to a class table that stores information about the classes and sections. Students were allowed to be duplicated if they had classes on multiple campuses, because these classes would not be able to overlap and would require the student to move their cars.
Looking at the graph, we can see a couple of patterns. We expect to see a bell like curve for parking demand because the majority of students are full time and are expected to be at school from morning until the afternoon. The flat line for the smaller campus is because there are few programs and almost all students have the same start and end times. The evening classes (even the ones out to 10pm!) are mainly early childhood education, parenting classes, and adult basic education classes.
Not included in this model is parking demand for staff and faculty, which is relatively predictable, and various in and out parking over the course of the day. It also does not account for mode split, which also was not available. The college likely has a high level of SOV drivers, especially at one of the suburban campuses off of I-5.
* This happens when a student is enrolled in a vocational program on one campus, but takes general education classes at the main campus.