Road black spot linear cause analysis method based on binary Logistic regression

Abstract

The invention discloses a road black spot linear cause analysis method based on binary Logistic regression, and the method comprises the following steps: 1, dividing a road to be researched into n road segment units according to the length; 2, analyzing the safety of each road segment unit, and determining road segments with the lower safety as black spots; 3, classifying the traffic accidents happening at each black spot into K classes according to the types of accidents, and converting the number of various types of accidents happening at the road segment units into a binary variable; 4, selecting Z types of linear indexes for describing the linear features of each black spot unit; 5, analyzing the impact on the occurrence of different types of accidents from various linear indexes of the black spots through a binary Logistic regression model. The method can be used for determining the impact on various types of accidents from the linear features of the road, helps a road traffic safety administrative department find a linear problem of a block spot, and is of the instructive significance to the design and construction of a road in future.

Description

technical field [0001] The invention belongs to the research field of road traffic safety management, and in particular relates to a method for analyzing the cause of road accident black spot section alignment based on binomial Logistic regression. Background technique [0002] The highway traffic system is a dynamic system composed of people, vehicles, roads, and the environment. The driver transmits the driving behavior to the vehicle by judging the environmental information, so that the movement of the vehicle adapts to the change of the road direction. Any imbalance in any link will have a negative impact on traffic safety. threaten. Among them, the impact of road alignment factors on traffic safety is most likely to be ignored by people, because there are few traffic accidents directly caused by road alignment factors, which mainly cause traffic accidents indirectly by affecting drivers or vehicles, and highway safety management departments are dealing with them. Accid...

AI Technical Summary

Problems solved by technology

[0004] The first type is based on historical traffic accident data, directly using the accident number method and accident rate method to identify accident black spots, or by constructing Poisson regression, negative binomial regression, and empirical Bayesian methods to regression analyze accident data. Judging accident black spots based on the prediction of accident development trend. The disadvantage of this type of method is that the indicators for identifying accident black spots are often relatively single, mainly using the number of accidents and the number of accident casualties as evaluation indicators, without considering road closures caused by traffic accidents. Indirect evaluation indicators such as time;

[0005] The second category is to use traffic conflict analysis technology (TCT) to judge possible traffic conflict points by predicting vehicle trajectories, and judge the location of more serious traffic conflicts as accident black spots. The traffic conflict analysis method relies on the historical data of traffic accidents. The degree is small, but it only considers the factors of traffic flow in traffic accidents, and cannot identify the black spots of accidents mainly caused by road conditions and environmental factors, and because the traffic conflict analysis method has a large workload, it is only applicable to small-scale cities Identification of Black Spots in Road Traffic Accidents

[0009] Although scholars at home and abroad have formed a relatively mature theoretical system in the analysis of the causes of traffic accidents, there are few special studies on the impact of road alignment characteristics on traffic accidents, especially the lack of parameters such as slope difference, slope difference change, flat curve radius, and transition curve length. Scientific Calculation Method in Highway Safety Design

Images