Spatiotemporal Influence of Land Use and Household Properties on Automobile Travel Demand

Xinyi Shen,Yujia Zhou,Sheng Jin,Dianhai Wang

DOI: https://doi.org/10.1016/j.trd.2020.102359

2020-01-01

Abstract:Understanding the patterns of automobile travel demand can help formulate policies to alleviate congestion and pollution. This study focuses on the influence of land use and household properties on automobile travel demand. Car license plate recognition (CLPR) data, point-of-interest (POI) data, and housing information data were utilized to obtain automobile travel demand along with the land use and household properties. A geographically and temporally weighted regression (GTWR) model was adopted to deal with both the spatial and temporal heterogeneity of travel demand. The spatial-temporal patterns of GTWR coefficients were analyzed. Also, comparative analyses were carried out between automobile and total person travel demand, and among travel demand of taxis, heavily-used private cars, and total automobiles. The results show that (I) The GTWR model has significantly higher accuracy compared with the Ordinary Least Square (OLS) model and the Geographically Weighted Regression (GWR) model, which means the GTWR model can measure both the spatial and temporal heterogeneity with high precision; (II) The influence of built environment and household properties on automobile travel demand varies with space and time. In particular, the temporal distribution of regression coefficients shows significant peak phenomenon; and (III) Comparative analyses indicate that residents' preference for automobiles over other travel modes varies with their travel purpose and destination. The above findings indicate that the proposed method can not only model spatial-temporal heterogeneous travel demand, but also provide a way to analyze the patterns of automobile travel demand.