Promote transit via hardening first-and-last-mile accessibility: Learned from modeling commuters' transit use

To understand how transit ridership would be increased via hardening first-and-last mile (F&LM) connections, the paper explores a quantity-based method to measure the effect of F&LM access when estimating workers' commuting transit use in Hamilton County, Ohio. Considering variations in the spatial distribution of socioeconomics, built environment, and transportation services, a geographically weighted lasso (GWL) based structure is applied in modeling transit use so as to capture spatial heterogeneous role of F&LM access and to deal with local variable multi-collinearity. In the GWL model, transit use is formulated as a function of transit service coverage, job accessibility by transit, and other socioeconomic, built environment, and transportation variables. The transit service coverage is articulated with choices of different F&LM modes (i.e., walking and biking), F&LM network connectivity to transit, and land use density; and job accessibility by transit is measured using the number of destination jobs covered by transit and transit journey time to destinations. The GWL modeling results suggest that a 10% growth in population covered by transit and job accessibility by transit can increase transit usage by 5.9% and 6.6% respectively, especially attracting more riders in the urban fringe. Then, the effectiveness of policy interventions such as dense development, promotion of bike usage, and increased bike connectivity in increasing transit ridership are subsequently estimated. Additionally, results indicate that providing improved F&LM access to car-less population, minority, and low-income people who tend to rely more on transit than others is more effective in promoting transit use.