Predicting Freeway Traffic Crash Severity Using XGBoost-Bayesian Network Model with Consideration of Features Interaction

Abstract

In the field of freeway traffic safety research, there is an increasing focus in studies on how to reduce the frequency and severity of traffic crashes. Although many studies divide factors into “human-vehicle-road-environment” and other dimensions to construct models whichshowthe characteristic patterns of each factor's influence on crash severity, there is still a lack of research on the interaction effect of road and environment characteristics on the severity of a freeway traffic crash. This research aims to explore the influence of road and environmental factors on the severity of a freeway traffic crash and establish a prediction model towards freeway traffic crash severity. Firstly, the obtained historical traffic crash data variables were screened, and 11 influencing factors were summarized from the perspective of road and environment, and the related variables were discretized. Furthermore, the XGBoost (eXtreme Gradient Boosting) model was established, and the SHAP (SHapley Additive exPlanation) value was introduced to interpret the XGBoost model; the importance ranking of the influence degree of each feature towards the target variables and the visualization of the global influence of each feature towards the target variables were both obtained. Then, the Bayesian network-based freeway traffic crash severity prediction model was constructed via the selected variables and their values, and the learning and prediction accuracy of the model were verified. Finally, based on the data of the case study, the prediction model was applied to predict the crash severity considering the interaction effect of various factors in road and environment dimensions. The results show that the characteristic variables of road side protection facility type (RSP), road section type (LAN), central isolation facility (CIF), lighting condition (LIG), and crash occurrence time (TIM) have significant effects on the traffic crash prediction model; the prediction performance of the model considering the interaction of road and environment is better than that of the model considering the influence of single condition; the prediction accuracy of XGBoost-Bayesian Network Model proposed in this research can reach 89.05%. The identification and prediction of traffic crash risk is a prerequisite for safety improvement, and the model proposed and results obtained in this research can provide a theoretical basis for related departments in freeway safety management.

A Parallel FP-Growth Mining Algorithm with Load Balancing Constraints for Traffic Crash Data

“…Most of the association rules extracted above include driver factors, so it indicates that human is the main operator in traffic and the key factor that determines whether a crash occurs or not. Some of the above results are consistent with those of some previous studies [1,7].…”

“…Traffic crash is a serious threat to public health safety and has become the eighth leading cause of human death. The number of deaths caused by traffic crashes continues to increase, with about 1.35 million people dying every year, and the main cause of death for people aged 5 to 29 is the traffic crash [1]. Especially on freeways, the vehicles usually run at a fast speed, once the traffic crash occurs, it is easier to cause serious casualties.…”

A Study on Pedestrian–Vehicle Conflict at Unsignalized Crosswalks Based on Game Theory

“…Ma et al [20] developed a pedestrian-vehicle conflict probability estimation model based on the arrival volume and operational characteristics of pedestrians and vehicles, by estimating the probability of a decision subject appearing on a crosswalk. In addition, some scholars classified the causes and levels of pedestrian-vehicle conflict by sequence comparison [21], and constructed a new evaluation system for quantitative analysis of the severity of traffic accidents [22], and for correlation analysis among variables [23] to determine the factors affecting pedestrian-vehicle conflict levels.…”

The Improved Precoding Method in the VLC-Based Intelligent Transportation System

“…To implement a robust VLC-based intelligent transportation system [19,20], the precoding method which is independent from channels/signals can realize data transmission reliably in various physical channel scenarios. e mapped DMT symbol is coded by multiplying the precoding matrix for Digital Signal Processing (DSP) at the DMTenabled transmitter.…”