Using Data Science for Equity at SFMTA
Student Capstone

Program Area(s):

Date: June 1, 2022

Author(s): Ryan Caro

Abstract

The San Francisco Municipal Transportation Agency (SFMTA) cut back service at the onset of the COVID-19 pandemic in 2020 and has been slowly rebuilding its service since. The SFMTA Board asks staff to conduct an equity analysis after each service change. To measure equity, SFMTA calculates the number of jobs that can be reached in a 30-, 45-, and 60-minute commute from specified “equity neighborhoods,” selected based on the percentage of households with low incomes, low rates of private vehicle ownership, and race and ethnicity
demographics.

My research is the first step in automating these analyses in order to build a tool to optimize SFMTA service for job access from equity neighborhoods. I use open-source tools to build a model of transit across the Bay Area and calculate the number of jobs available within the specified commute times from each equity neighborhood.

I find that although this tool is planned to help SFMTA improve its service, proximity to BART is the greatest predictor of job access within 45 and 60 minutes. I find that among open-source tools, OpenTripPlanner’s point-to-point methods are significantly more accurate than its one-to-many methods, despite being several orders of magnitude slower. I also make recommendations for how SFMTA can continue to build this tool. I find that equity neighborhoods are a very rough proxy for disadvantaged groups, and I suggest that SFMTA create an equity index to weight tracts across the entire city rather than focusing exclusively on equity neighborhoods.

About the Project

This capstone project was completed in partial satisfaction of the requirements for the degree Master of Urban & Regional Planning at the UCLA Luskin School of Public Affairs.