Study On The Influencing Factors Of Crossing Dangerous Behaviors Of Takeaway Electric Bicycles At Signalized Intersections

The rapid development of “Internet + Catering” has spawned and expanded the food delivery industry,bringing great convenience to people’s life.But at the same time,traffic violations by takeaway riders during the delivery of food are very common,resulting in a high incidence of traffic accidents during delivery.As an important node of the urban road network,the signalized intersection is the main place where urban road traffic conflicts and traffic accidents occur.At present,there are very few research results on the crossing safety of takeaway electric bicycles at signal intersections at home and abroad,but traffic violations of takeaway electric bicycles at signalized intersections in major cities in China are common and accidents occur frequently.Therefore,this paper takes the takeaway electric bicycle as the research object,and focuses on the analysis of the dangerous behavior and its influencing factors during the process of the takeaway electric bicycle crossing the signalized intersection.The main work of this paper is as follows:(1)Basic characteristics and accident analysis of takeaway electric bicycles.The detailed analysis of the basic characteristics of takeaway electric bicycles from three aspects of vehicle characteristics,use characteristics and traffic characteristics.Then it introduces the traffic accident status of takeaway electric bicycles in some cities in China and analyzes the cause of the accident.(2)Empirical research on the crossing behavior of takeaway electric bicycles.The crossing behavior of takeaway electric bicycles at typical signalized intersections in cities was selected for field shooting to obtain basic data and conduct data processing.Analyze the waiting position distribution,waiting time distribution,movement trajectory and secondary crossing of takeaway electric bicycles in the process of crossing the signalized intersection,and reveals the microscopic behavior distribution of takeaway electric bicycles at the signalized intersection.(3)Modeling and analysis of dangerous behavior of takeaway electric bicycles crossing at signalized intersections.Logistic regression model and XGBoost algorithm were used to establish the influencing factor model of dangerous behavior of takeaway electric bicycles at signalized intersection.The classification effects of the two models are compared and analyzed.The results show that the classification effect of XGBoost is better than that of Logistic regression model.Then analyze the influencing factors based on the results of the XGBoost model and put forward targeted safety improvement suggestions.(4)Analysis of waiting time of takeaway riders at signalized intersection.Using the Kaplan-Meier analysis to compare and analyze the overall distribution of waiting time of takeaway riders when the data was deleted or not,and to compare and analyze the differences of waiting time of takeaway riders under the influence of single factor.Then Cox proportional risk regression model was used to analyze the influencing factors of the waiting time of takeaway riders at signalized intersection.The research in this paper can enrich the existing research,and has important theoretical and practical significance for preventing and reducing the dangerous behavior of takeaway electric bicycles and improving the safety level of signalized intersections.