Establishing the relationship between built environment and traffic congestion states provides a feasible direction to explore factors affecting the spatial distribution of road traffic states, and supports to meet needs of traffic management and control under the information environment. The study analyzes the spatial pattern of traffic state, establishes two regression models to identify factors affecting local traffic congestion, and visualizes the impact degree of spatial heterogeneity. Firstly, the real data are collected and processed, and congested grids are selected as study range. Then, independent and dependent variables are obtained, presented in grids by ArcGis. Third, the ordinary least squares (OLS) and the geographically weighted regression (GWR) are applied to fitting. Results show the influence and the influential degree are different in spatial distribution. Analysis of the traffic state by GWR based on grids is feasible to present the traffic state.

traffic state;built environment;spatial heterogeneity;geographically weighted regression