考虑车道变换影响的高速公路交通事故预测模型研究

发布时间：2018-04-13 04:06

本文选题：交通工程 + 事故预测模型　；参考：《公路交通科技》2014年06期

【摘要】：考虑车道变换可能对交通安全造成不利影响,结合广东省3条高速公路64个路段的交通运行状况数据和交通事故历史数据,利用负二项分布预测方法,建立并标定了基于交通量、路段长度、车道变换次数、大型车变道比例、单位里程变道次数等5个解释变量10组不同组合的交通事故预测模型。通过计算各组模型的Akaike信息量准则指标,得到了3组权衡了模型结构(即解释变量数量)和数据拟合度的最优模型。结果表明,虽然3组最优预测模型的预测精度仍有待提高,但是考虑车道变换影响的交通事故预测模型明显优于其他模型。这说明与车道变换相关的变量可以作为交通事故预测的有效解释变量,并且引入该类型变量可以更好地预测高速公路交通事故的发生。
[Abstract]:Considering the possible adverse impact of lane change on traffic safety, combined with traffic operation status data and traffic accident history data of 64 sections of three expressways in Guangdong Province, a negative binomial distribution prediction method is used.The traffic accident prediction model was established and calibrated based on five explanatory variables, such as traffic volume, length of section, number of lane changes, ratio of large vehicle to lane change, and number of road changes per unit mileage, and 10 groups of different combinations of traffic accident prediction models were established and calibrated.By calculating the Akaike information quantity criterion of each group of models, three groups of optimal models are obtained, which weigh the structure of the model (that is, the number of explanatory variables) and the data fit degree.The results show that although the prediction accuracy of the three groups of optimal prediction models is still to be improved, the traffic accident prediction model considering the influence of lane change is obviously superior to other models.This indicates that the variable related to lane change can be used as an effective explanatory variable for traffic accident prediction, and the introduction of this type of variable can better predict the occurrence of expressway traffic accident.
【作者单位】：哈尔滨工业大学交通科学与工程学院;
【基金】：广东省交通运输厅科技项目(2012-2013)
【分类号】：U491.31

【参考文献】