Developing Real-Life Measures of Transit Accessibility

Abstract: This research paper provides a quantitative methodology for measuring transportation accessibility. The word “accessibility” is used in a broad sense and includes multiple measures that will help one assess the efficiency of a transportation system. The findings in this paper specifically represent a case study on transit accessibility both at the state level and the city level.

Instead of manually calculating directions for thousands of trips, Python coding in conjunction with APIs (application programming interfaces) was used. First, millions of addresses were compiled, the data was cleaned, and a percentage of records was randomly chosen. The addresses were then geocoded and destinations within the vicinity were chosen. Next, transit directions were retrieved using a Google API, and the accessibility measures for each trip were calculated. Additionally, driving directions were retrieved to compare the transit travel time to the driving time. Finally, the results were mapped in ArcGIS to visualize each measure and its implications.

The results were expected to not be favorable concerning transit, but this research provided the data to provide quantitative results. Many of the findings were worse than expected. The most significant measure was the transit to driving travel time ratio. This ratio yielded a factor showing how much longer transit would take compared to driving. The minimum average ratio for a county in California was greater than two, and the minimum average ratio for a block group in Fresno City was greater than three. Based on these findings, it is unlikely that the public are willing to transition from using their vehicles to using transit.