Assessing Use of Pedestrian Crash Data to Identify Unsafe Transit Service Segments for Safety Improvements

Abstract

This paper assesses the use of pedestrian crash data to explain the role of transit service and ridership in pedestrian crashes in urban areas. The values for measures that may be used to identify segments with transit service where it is potentially unsafe for users to walk were computed to identify the need for safety improvements. Descriptive analysis was conducted by comparing measures such as the numbers of pedestrian crashes, pedestrian crashes per mile, pedestrian crashes per mile per 10 transit stops, pedestrian crashes per mile per 1,000 transit riders (boarding and alighting), and pedestrian crashes per million vehicle miles traveled (MVMT), using data for Charlotte, North Carolina. Thirty study segments with transit service and 30 segments without transit service were considered to compare and examine the relationship. These segments were selected such that the land uses and on-network characteristics were similar and comparable. The computed ratios of the mean values of the numbers of pedestrian crashes, pedestrian crashes per mile, and pedestrian crashes per MVMT on segments with and without transit service were 4.82, 4.92, and 2.70, respectively. Statistical analysis was also conducted to test if the values of the selected measures were significantly higher on segments with transit service than on segments without transit service. Results from both descriptive and statistical analyses support the general belief that pedestrian crash rates are high on transit corridors and that a significant number of these pedestrian crashes involve transit system riders. The values for measures based on pedestrian crash data are thus computed, compared, and combined for segments with transit service to identify segments potentially unsafe for transit system users.