Assessing the Impact of Streetscape on Shared Bike Ridership with Street View Imagery

Keywords: Bike-sharing, GIS, Street View Imaginary, Urban Planning, Streetscape
Session Type: Virtual Paper Abstract
Day: Tuesday
Session Start / End Time: 3/1/2022 09:40 AM (Eastern Time (US & Canada)) - 3/1/2022 11:00 AM (Eastern Time (US & Canada))
Room: Virtual 14

Abstract

The automobile's over-reliance has resulted in a slew of urban problems in the United States, including congestion, traffic injuries, and urban sprawl. In recent years, the advent of bike-sharing systems has allowed people to use shared bikes for short trips or to complete the first or last mile. However, because the bicycling environment and infrastructure vary each city, the adoption of bike-sharing systems varies as well. While earlier research has concentrated on the built environment and its effect on shared bike use, very few studies have explored street elements. This is because quantifying street elements is difficult and data collection is costly and time consuming. The recent availability of Street View imagery, as well as advancements in computer vision, enables quick data collection and quantification of street elements. Expanding on prior research, the purpose of this study is to determine the effect of streetscape on shared bike ridership while controlling for land-use and built environment effects. Street elements were extracted from Google Street View images by a deep learning model and associated with Boston's shared bike data through negative binomial regression model. Consistent with earlier research, the findings indicate that the characteristics of land use and the built environment have a substantial impact on shared bike ridership. Additionally, this research reveals that the streetscape influences shared bike ridership. For instance, street greenery and openness are associated with increased shared bike riding. This finding can aid urban planners and designers in developing more effective strategies for increasing bike-sharing system adoption.