Yu Jin,Xuesong Wang,Xiaohong Chen
Abstract:Traffic safety has received increasing attention. Many macro level safety models explore the relationship between crash occurrence and explanatory variables. Although road network pattern is an essential aspect for transportation planning, studies of its safety effect are limited. In this paper, the meshedness coefficient was used to mirror the network structure and examine its effect on traffic analysis zone (TAZ) level safety. Data of 662 TAZs from Orange County, Florida, U.S.A. were collected. A conditional autoregressive model which considers the spatial correlations among TAZs was developed. Estimation results showed the meshedness coefficient performed well in capturing the nature of network patterns, as well as in building relationship with zonal level crashes. This study indicates that, besides traditional network size variables such as road length, structure of road network should also be considered in developing more precise zonal safety prediction models.