基于EWM-FCE模型的高速公路行车风险分析

郭懿天¹，张萌萌^1*，潘景宜²

1.山东交通学院交通与物流工程学院，山东 济南  250357；

2.齐鲁工业大学(山东省科学院)经济与管理学部，山东 济南  250353

摘要：为准确评估高速公路行车风险，综合考虑微观驾驶行为和宏观交通环境，选取速度离散度、加速度变化率、碰撞时间占比、空间占有率及中大型车占比为行车风险的评价指标，提出基于熵权法（entropy weighting method，EWM）与模糊综合评价（fuzzy comprehensive evaluation，FCE）的EWM-FCE高速公路行车风险评价模型。通过问卷调查数据标定EWM-FCE模型的组合隶属度函数，以山东省某高速公路路段为例，评价此路段的行车风险，并将其评价结果与传统三角形隶属度函数及梯形隶属度函数进行对比，验证EWM-FCE模型的准确性。结果表明：该高速公路路段的行车风险等级为超低风险；采用组合隶属度函数的EWM-FCE模型相较于采用三角形隶属度函数及梯形隶属度函数时，评价结果准确率分别提高7.5%和7.6%。

关键词：行车风险分析；FCE；评价指标体系；组合隶属度函数；熵权法

Highway traffic risk analysis based on EWM-FCE model

GUO Yitian¹，ZHANG Mengmeng^1*，PAN Jingyi²

1.School of Transportation and Logistics Engineering, Shandong Jiaotong University, Jinan 250357, China;

2.Faculty of Economics and Management, Qilu University of Technology(Shandong Academy of Sciences), Jinan 250353, China

Abstract: In order to accurately evaluate the risk of highway driving, considering micro driving behavior and macro traffic environment, speed dispersion, acceleration change rate, collision time proportion, space occupancy rate, and proportion of medium and large vehicles are selected as evaluation indicators of driving risk. The EWM-FCE model based on entropy weighting method(EWM) and fuzzy comprehensive evaluation (FCE) is proposed. The combined membership function of the model is calibrated through questionnaire survey data, taking a certain highway section in Shandong Province as an example, to evaluate the driving risk of this section, and the evaluation results are compared with the traditional triangular membership function and trapezoidal membership function to verify the accuracy of the EWM-FCE model. The results show that the driving risk level of this highway section is ultra-low risk; the EWM-FCE model using the combined membership function improves the accuracy of the evaluation results by 7.5% and 7.6%, respectively, compared with the triangular and trapezoidal membership functions.

Keywords: driving risk analysis; FCE; evaluation indicator system; combined membership function; EWM