Division and Analysis of Accident-Prone Areas near Highway Ramps Based on Spatial Autocorrelation

Abstract

This study focuses on identifying accident-prone areas and analyzing the factors contributing to the distribution of traffic accidents near highway ramps. A combined method of kernel density estimation, spatial autocorrelation analysis, and multivariate logistic regression analysis helped to identify accident hotspots. Through data collection and analysis, the clustering characteristics of traffic accidents in the diversion and merging areas were identified. Four levels of accident-prone areas were divided according to the accident rates. The factors influencing the spatial distribution of accidents were analyzed. The results showed that traffic accidents in the diversion area were concentrated near the exit, but the accidents in merging areas had a wider range of distribution. The analysis of this phenomenon was conducted using the multinomial logit model results. The important factors of different accident-prone areas were clarified. The temperature, the accident lane, weather conditions, and the time of day had significant impacts on the spatial distribution of traffic accidents. The study’s findings provide an important decision-making basis for highway accident prevention management.