Study of the Risk of Truck Rear-End Collision in a Continuous Downhill Section of an Expressway

Abstract

To improve the safety of truck traffic on continuous downhill sections of highways, optimize traffic safety management and the design of traffic safety facilities, study in-depth the risk of truck rear-end accidents on continuous downhill sections of highways, based on actual measurement data, this paper puts forward the evaluation index of rear-end risk considering the speed difference and headway time distance called the Collision Deceleration Rate (CDR). First, the principle of risk of truck rear-end accidents on long downhill road sections was analyzed, and the data of trucks travelling on long downhill road sections in a special control area of a western motorway were used to determine the risk-influencing factors as speed difference and headway time distance; then, the distribution law of effective CDR data was analyzed and calculated through a fitting distribution test, and the 85% quantile value of its distribution was taken as the threshold value of dangerous CDR; finally, the collision potential index (CPI) is combined with the concept (CPI) to establish a CDR-based continuous downhill truck rear-end risk probability model. Based on the tailgating accident data in the special control area of the motorway, the CDR-based continuous downhill truck tailgating risk probability model was applied and tested, and the results showed that: in the fitted distribution test, the CDR samples of the studied continuous downhill section obeyed lognormal distribution, and the threshold value of hazardous CDR is 0.4916 m/s2. There are more significant differences in the tailgating probability risk at different measurement points, while the tailgating risk are different. The correlation coefficient between the accident rate data and the probability risk data is r(X,Y)=94.5%, which is a strong correlation, i.e. the trend of the accident rate data and the probability risk data is consistent, indicating that the CDR-based probability model of tailgating accident risk can provide a scientific assessment basis for the risk study of continuous downhill sections of highways, and the research method can be used as a reference for other sections of highways.