Spatial regression modeling of bus ridership using smart card data and various built environmental features

Bus boarding ridership is a key indicator for public transportation efficiency.Increased ridership can help reduce traffic congestion and positively affect the environment.This study aims to examine the spatial relationship between bus boarding ridership and the built environment, utilizing smart card and point of interest (POI) data.The research employs methods including ordinary least squares (OLS), spatial lag model (SLM), and geographically weighted regression (GWR) for comprehensive spatial analysis.A stepwise OLS model was employed to identify variables as inputs in the other models.The findings suggest that the GWR model provides a better understanding of the factors that affect bus boarding ridership.