| Urban expressway is the highest-grade road in the urban road network.It has a large traffic carrying capacity and fast vehicle traffic speed.It is an important part of the urban road network.However,with the rapid development of urban economy and infrastructure,the traffic flow on expressways is also increasing day by day,and causing more traffic accidents.Under the condition of mixed traffic flow,the speed difference between vehicles due to the difference of vehicle types is an important factor affecting traffic safety.Therefore,it is important to analyze the impact of different proportions of trucks in mixed traffic flow on traffic safety according to whether there is restriction for trucks on expressway.Therefore,focusing on traffic safety,exploring the correlation between the frequency of urban expressway accidents and the proportion of trucks,developing a prediction model for urban expressway accidents,and analyzing the safety impact of different proportions of trucks is the key point of this research.Relevant studies have shown that the mixed traffic flow of trucks and general passenger vehicles will lead to a decrease in the level of traffic safety,but there is still a lack of systematic analysis of the impact of different truck proportion on the traffic safety of urban expressways.Therefore,this paper aggregates the accident prediction data according to the proportion of trucks,introduces parameters such as the proportion of trucks into the accident prediction model Safety Performance Function(SPF)model,and develops urban expressway accident prediction considering the proportion of trucks,using the methods of marginal effect,crash modification factor and simulation evaluation to analyze the impact of the proportion of trucks on the traffic safety of urban expressways.The research work of this paper mainly includes the following contents:Firstly,starting from the obtained traffic flow and historical accident data,the data is preprocessed.It mainly includes calculating the traffic flow parameters of the road section from the passing data of the checkpoint,such as the annual average daily traffic volume,the proportion of trucks.According to the description of the accident location contained in the accident information to determine the latitude and longitude coordinates of the accident location.Useing the correlation analysis method to verify whether there is a correlation between the proportion of trucks and the number of accidents,and useing statistical methods to analyze the characteristics of historical accidents and traffic flow data to support subsequent accident prediction models construction that consider the proportion of trucks.Secondly,based on the high-precision map and geographic information system software,the road segmentation is completed according to the road geometric characteristics,and the processed historical traffic flow data and accident data are aggregated in the two dimensions of space and time according to the road segmentation results.The reliability of the model parameter fitting results,the outliers were removed from the data set,and the data were aggregated for four types of accidents: singlevehicle simple accident,multi-vehicle simple accident,single-vehicle general accident and multi-vehicle general accident according to the accident form and severity.Complete the construction of the accident prediction dataset.Thirdly,based on the accident prediction data set,the annual accident frequency of the road segment is used as the dependent variable,the number of lanes and road segment types are used as factors,and parameters such as road segment length,annual average daily traffic(AADT)and truck proportion are used as covariates,and negative binomial regression and zero inflation binomial regression are used.Two regression methods are used to fit the parameters of the urban expressway accident prediction model considering the proportion of trucks,and the prediction effect and goodness of fit of the model are evaluated by indicators such as root mean square error and loglikelihood value.The p-value of the evaluation results is less than 0.05,indicating that at 95% confidence,the regression results of the prediction model parameters are significant.Finally,based on the development of an urban expressway accident prediction model considering the proportion of trucks,the impact of the proportion of trucks on the traffic safety of urban expressways is analyzed.Based on the fitting results of the parameters of the accident prediction model considering the proportion of trucks,the marginal effect analysis method is used to calculate the elastic coefficient and the predicted marginal value of the proportion of trucks in the interval of [0,0.2),and the crosssectional analysis method is used to calculate the crash modification factor(CMF)caused by the proportion of trucks.The CMF is used to quantitatively analyze the impact of the proportion of trucks on the frequency of accidents.The regression effects and prediction accuracy of the urban expressway accident prediction model considering the proportion of trucks and the basic accident frequency prediction model considering the CMF of the proportion of trucks are compared and analyzed.Using micro-traffic simulation and Surrogate Safety Assessment Model(SSAM)to analyze the correlation between the traffic safety level and the proportion of trucks in the interval of [0,1],the research results show that the urban expressway accident prediction model developed in this paper can effectively predict the accident frequency of the urban expression segments.In the interval [0,0.2),the introduction of trucks will slightly reduce the traffic safety level of urban expressways;the microscopic traffic simulation experiment results show that in the interval [0.2,0.8),with the increase of the proportion of trucks,the level of traffic safety gradually declines In the [0.8,1] interval,as the proportion of trucks gradually approaches 1,the traffic safety level begins to rise,but is still lower than the situation where the proportion of trucks is 0.In summary,this study is based on the analysis of real accident data and the construction of accident prediction models,as well as the analysis results of the impact of trucks on traffic safety based on actual data and simulation evaluation.It provides an important theoretical basis for the management of expressway traffic restriction. |
