Inspecting urban transit-oriented development from the perspective of human activity: A case study of Xi'an, China

Evaluation of transit-oriented development (TOD) can inform the exploitation of transportation and land use around metro stations. This study evaluates TOD by focusing on human activity patterns around metro stations, addressing a gap in previous research that mainly integrated transport capacity, land use, and the built environment. Using data from Xi'an, China, three types of human flow (home-work, home-other, work-other) are analyzed to construct five-dimensional activity indicators representing diversity, intensity, and the proportion of home, work, and other activities near metro stations. Stations are classified into seven clusters based on these activity characteristics. Multinomial logistic regression and random forest models are applied to examine the relationship between the built environment and TOD clusters, revealing their impact on human activity patterns. The findings offer insights into how the built environment influences TOD categories, supporting future transportation planning and decision-making.