Research On Identification And Causality Analysis Methods Of Traffic Accident Hot Zones Based On ArcGIS

With the continuous increasing of road total mileage,the problem of traffic safety has intensified,and it has become a bottleneck problem that restricts the sustainable development of the social economy.In order to improve traffic safety,formulate targeted measures and suggestions for improvement,it is important to explore the spatial distribution of accident hot zones and understand the causes of traffic accidents.Traditional traffic accident hotspot identification method cannot effectively identify continuous dangerous sections in terms of accident locations.In addition,although the analysis of the causes of traffic accidents is one of the research hotspots in the field of road traffic safety,few researchers have explored it from the perspective of road special environment.Based on the above situation,this paper initially proposes a new method based on geographic information system(GIS)with taking the spatial autocorrelation between accident locations into consideration,which can help explore the spatial distribution characteristics of road traffic accidents and identify areas with high incidence of accidents.Specifically,it includes the application of GIS information platform to achieve the spatial data storage of road traffic accidents,query and extraction traffic accidents and road environment related information,taking into account the spatial autocorrelation between traffic accident points,find out the continuous neighborhoods with high traffic accident risk index in the road network by the method of classic Local Moran’s index algorithm.Then establish a causality evaluation method based on Logistic theory,and make an overall design,including the establishment of a logical theory-based causality analysis method,the determination of the basic unit and scale of regression analysis,data transformation ideas and detailed steps of the spatial information.Then,the detailed operating steps of causal analysis in ArcGIS and SPSS platform were introduced and the program was implemented.Finally,the city of Enschede,the Netherlands was selected as a study example to verify the validity of the identification method of the accident hot zones,whose accuracy was proved through the environment audit.Besides,the reliability level of regression model was proved by the detection indexes including the significance and the goodness fit of the model.The research results of this paper can realize the effective transformation of accident records and the identification and visualization of accident hotspots by combining the accidents attribution and accident location attribution together.It is not only clearly explaining high-risk accident areas to professionals,but also can be intuitively understood by non-professionals.In addition,the impact analysis of road environmental factors on the distribution of accident hotspots has profound implications for assisting road designers in formulating safety-related and sustainable road development-related strategies and improving the safety of municipal roads.