Evaluation On The Safety Of Freeway Merging Areas Considering Driver’s Errors

Merging areas are important for freeways because they connect the main roads and on-ramps.In freeway merging areas,due to the complex geometric configuration and the changeable traffic conditions,traffic accidents are likely to occur once the driver makes an error.In previous studies,there is a lack of research on the safety of freeway merging areas considering road conditions,traffic flow characteristics and driver behaviors.This study has explored the safety of freeway merging areas considering the driver’s errors.The main research contents and methods are as follows:First,the research status quo of freeway merging areas in the world has been summarized;then main factors influencing the traffic safety of merging areas were preliminarily determined.Next,the traffic charac teristics of merging areas were analyzed.Based on accident information(such as accident types and causes),the relationship between the types of accidents in interchange ramps and the driver’s errors was analyzed,which serves as a theoretical basis for determining the influencing factors of safety in merging areas.With respect to the different construction forms of freeway merging areas,simulation models in the forms of single-lane and two-lane ramps in merging areas were established respectively.Relevant parameters for the models were calibrated with the aid of the Genetic Algorithm,which can ensure the error between the output value and the target value of the models is below 5% so that the simulation models can effectively represent the actual traffic conditions.408 groups of traffic schemes were designed through orthogonal experiment.In the simulation models,two error types(the driver’s perception error and judgment error)were simulated;and their traffic conflict data were output respectively to establish a database for traffic safety evaluation.The safety risk for each traffic scheme was evaluated through the grey clustering method.Based on the results of safety risk assessment,prediction models for safety risk were established for the merging areas of the freeway main roads and the single-lane on-ramps and two-lane on-ramps,respectively.Next,the effectiveness of the safety risk prediction models was verified with reference to the data of the past 10-year accidents in Pennsylvania.The results reveal that the correlation coefficient between the risk ranking of the prediction models and the ranking based on the accident data is0.9,indicating a strong positive correlation.The risk of the merging areas in the supporting project was evaluated by the safety risk prediction models;then the predicted risk value and risk grade of each merging area were obtained.According to the predicted risk,an improvement measure to lengthen the on-ramp markings in some merging areas was put forward.The safety risk prediction models in the forms of one-lane and two-lane ramps proposed in this study can improve the safety evaluation theory of highway projects,and provide theoretical support for safety risk evaluat ion and traffic dynamic management and control in engineering designs,which is of practical engineering significance.