Understanding the time-dependent effect of built environment attributes on station-level metro ridership uncertainty in Beijing: A big data analytic approach

Abstract

The relationship between built environment and metro (underground railway) ridership has been well discussed in the literature. However, few studies focus on the metro ridership uncertainty and its temporal association with the built environment, that is the time-dependent effect of built environment on metro ridership uncertainty. To fill this gap, this study investigates the spatio-temporal patterns of metro ridership uncertainty using data collected in Beijing. Multivariate analysis models are developed to establish statistical relations between metro ridership uncertainty and built environment for different time periods (day of the week and time of the day) for analyzing the temporal variations of the built environment impacts. Results show that time-dependent effects significantly exist, with hourly variations in the relationship more evident than daily variations. Stations with added employment and residential buildings would generate additional metro demand during weekdays, while stations with added service and commercial establishments are during weekends. The findings of this study will help us understand how the built environment affects metro ridership uncertainty in different time periods and develop effective policies.