Research On The Traffic Conflict And Safety Evaluation Of Freeway Typical Sections

Though the road traffic safety evaluation based on historical crash data has beendeveloped well, it has a limited development as it’s hard to get the crash data or the datacollected is incomplete. Another reason makes this method cannot be used effectively isthat new road doesn’t operate long makes the crash data can’t meet the statisticalrequirements. So many researchers try to use none crash data method, also calledsurrogate safety measures, on road safety research, and traffic conflict technique (TCT)is one mature theory of them. The traffic conflict on freeway has fewer actors and clearconflict types than low grade highway or urban road, and the continuous traffic flow onfreeway is suitable for the application of TCT. For these reasons, normal horizontal andvertical alignment combination section, work zone and merging area are selected astypical freeway section; the TCT and the safety study of typical freeway sections basedon TCT are studied in this paper.Firstly, the feasibility and advantage of freeway traffic safety study based on trafficconflict technique are analyzed and summarized, and the traffic conflict definition oftypical freeway sections is proposed. Then the data acquisition and extraction schemeare formulated after each typical section are selected. And what’s more, the traffic flowcharacteristics and safety feature of each section are analyzed.Secondly, start with finding the traffic conflict indicators calculation modelsmeasured by time, space and others. Combine the traffic data surveyed; choose theindicators from the point of road section and individual vehicle or called as macroscopicand microcosmic. So the time to collision (TTC) model and deceleration rate to avoidcrash (DRAC) model are selected and modified to calculate the traffic conflict rate andrisk of each survey sections. Lastly, the side crash conflict calculation model is builtbased on TTC indicator according to research needs, and the serious conflict measurestandard is determined.After the calculation of conflict indicators, the factors may have a highlightinfluence on conflict are preliminary screened by studying the accident influence factors.Then the macro and micro dominant influence factors of each section are programmedscreened by multiple regression method. The relationship between every factors andconflict rate are found by control variables method also. The K-means clusteringalgorithm is used to divide the traffic safety levels of study sections in to four categories.So their safety conditions are evaluated.At last, slope bottom into horizontal curve at downhill direction in normalhorizontal and vertical alignment combination section are regarded as a potentialdangerous road sections, and the traffic safety precautions are given to the point by theirprominent influence factors. Then, simulation models are built based on the section had been laid safety guarantee facilities. And the safety level of the section aftermanagement is evaluated by using the above method based on the simulation data. Sothe effectiveness of the traffic safety precautions is verified.Research shows that the traffic safety evaluation based on TCT is a practicalmethod, because it can get a reasonable result by few fields of inquiry in a short time,and it has a wide prospect of application.