融合事故类型与频次的高速公路交通风险评估方法研究
Research on Highway Traffic Risk Assessment Method Integrating Accident Types and Frequencies
DOI: 10.12677/ojtt.2026.152022, PDF,    科研立项经费支持
作者: 崔 建, 孙砚涛, 李 璇, 白雪倩:山东高速股份有限公司,山东 济南;刘 凯*, 孟依冉:山东交通学院交通与物流工程学院,山东 济南
关键词: 交通安全高速公路风险评估熵权法K-Means聚类Traffic Safety Expressway Risk Assessment Entropy Weight Method K-Means Clustering
摘要: 为评估高速公路交通事故风险特征,本研究提出一种融合事故类型风险程度与频次的风险评估方法,并以某省事故数据为例分析其时空特征。首先,构建涵盖事故清障时间、清障类型、车辆数目以及类型的评估指标体系,采用熵权法确定不同事故类型(如侧翻、追尾等)的风险系数;其次,以单位路段或时间长度为基准,基于不同交通事故类型风险系数和频次,采用线性加权方法确定综合风险值;最后,采用K-means聚类方法划分风险等级,探究时空分布规律。结果表明,侧翻事故的风险最高(0.735),追尾(0.657)和自燃事故(0.540)次之;空间上,弯道中高风险比例高于直道,坡度越大高风险占比越低;时间上,白天的交通事故风险高于夜间和凌晨。该研究方法以易获取的事故处理和车辆数据为基础,为高速公路运营部门提供了一种高效可行的风险识别工具,有助于提升交通安全管理决策水平。
Abstract: To assess the risk characteristics of expressway traffic accidents, this study proposes a risk assessment method that integrates the risk degree and frequency of accident types, and takes the accident data of a certain province as an example to analyze its spatio-temporal characteristics. Firstly, an evaluation index system covering accident clearance time, clearance type, number of vehicles and types is constructed. The entropy weight method is adopted to determine the risk coefficients of different accident types (such as rollover, rear-end collision, etc.). Secondly, taking the unit road section or time length as the benchmark, based on the risk coefficients and frequencies of different types of traffic accidents, the comprehensive risk value is determined by using the linear weighting method. Finally, the K-means clustering method is adopted to classify the risk levels and explore the spatio-temporal distribution patterns. The results show that the risk of rollover accidents is the highest (0.735), followed by rear-end collisions (0.657) and spontaneous combustion accidents (0.540). Spatially, the proportion of high-risk sections in curves is higher than that in straight sections, and the greater the slope, the lower the proportion of high-risk sections. In terms of time, the risk of traffic accidents during the day is higher than that at night and in the early hours of the morning. This research method, based on easily accessible accident handling and vehicle data, provides an efficient and feasible risk identification tool for highway operation departments, which is conducive to improving the decision-making level of traffic safety management.
文章引用:崔建, 孙砚涛, 李璇, 白雪倩, 刘凯, 孟依冉. 融合事故类型与频次的高速公路交通风险评估方法研究[J]. 交通技术, 2026, 15(2): 239-247. https://doi.org/10.12677/ojtt.2026.152022

参考文献

[1] 张长生, 马荣国. 山区高速公路交通事故分析及多发路段鉴别[J]. 长安大学学报(自然科学版), 2010, 30(6): 76-80.
[2] Golov, E., Evtyukov, S., Protsuto, M., Evtyukov, S. and Sorokina, E. (2022) Influence of the Road Surface Roughness (According to the International Roughness Index) on Road Safety. Transportation Research Procedia, 63, 999-1006. [Google Scholar] [CrossRef
[3] 鲍彦莅. 基于交通环境的车辆风险分析及在途车速预警研究[D]: [硕士学位论文]. 昆明: 昆明理工大学, 2019.
[4] Iqbal, A., Rehman, Z.U., Ali, S., Ullah, K. and Ghani, U. (2020) Road Traffic Accident Analysis and Identification of Black Spot Locations on Highway. Civil Engineering Journal, 6, 2448-2456. [Google Scholar] [CrossRef
[5] 赵敏. 高速公路长大纵坡路段交通安全风险评估与分析[J]. 山西交通科技, 2023(1): 107-110.
[6] 贾洪飞, 桑珩, 杨丽丽, 等. 新建高速公路潜在事故风险点(段)识别方法[J]. 北京工业大学学报, 2016, 42(8): 1233-1238.
[7] 杨妮. 基于多源数据的高速公路路段风险判别[D]: [硕士学位论文]. 昆明: 昆明理工大学, 2023.
[8] Wang, D., Huang, Y. and Cai, Z. (2023) A Two-Phase Clustering Approach for Traffic Accident Black Spots Identification: Integrated Gis-Based Processing and HDBSCAN Model. International Journal of Injury Control and Safety Promotion, 30, 270-281. [Google Scholar] [CrossRef] [PubMed]
[9] Rahman, M.A., Moomen, M., Khan, W.A. and Codjoe, J. (2024) An Analysis of the Impact of Injury Severity on Incident Clearance Time on Urban Interstates Using a Bivariate Random-Parameter Probit Model. Stats, 7, 863-874. [Google Scholar] [CrossRef
[10] Li, R., Pereira, F.C. and Ben-Akiva, M.E. (2015) Competing Risks Mixture Model for Traffic Incident Duration Prediction. Accident Analysis & Prevention, 75, 192-201. [Google Scholar] [CrossRef] [PubMed]
[11] Jeon, H., Kim, J., Moon, Y. and Park, J. (2021) Factors Affecting Injury Severity and the Number of Vehicles Involved in a Freeway Traffic Accident: Investigating Their Heterogeneous Effects by Facility Type Using a Latent Class Approach. International Journal of Injury Control and Safety Promotion, 28, 521-530. [Google Scholar] [CrossRef] [PubMed]
[12] Drakopoulos, A., Shrestha, M. and Örnek, E. (2001) Freeway Crash Timeline Characteristics and Uses. Transportation Research Record: Journal of the Transportation Research Board, 1748, 132-143. [Google Scholar] [CrossRef
[13] Lee, S., Park, J. and Abdel-Aty, M. (2024) Hazards-based Duration Time Model with Priorities Considering Unobserved Heterogeneity Using Real-Time Traffic and Weather Big Data. Transportation Research Record: Journal of the Transportation Research Board, 2679, 1695-1707. [Google Scholar] [CrossRef
[14] 戢晓峰, 吴亚欣, 袁华智, 等. 二级公路交通流特征对事故严重程度的影响[J]. 中国公路学报, 2020, 33(2): 135-145.
[15] 戴学臻, 胡传喜, 陈茜, 等. 涉事车辆分类的高速公路重特大交通事故成因与严重程度研究[J]. 安全与环境学报, 2022, 22(5): 2583-2591.
[16] 温惠英, 马肇良, 赵胜, 等. 山区高速公路货车事故影响因素分析[J]. 华南理工大学学报(自然科学版), 2025, 53(7): 93-103.
[17] 乔建刚, 柯秋雨. 基于灰色关联分析和熵权法的山区干线公路事故影响因素研究[J]. 河北工业大学学报, 2020, 49(3): 81-85.
[18] McQueen, J.B. (1967) Some Methods of Classification and Analysis of Multivariate Observations. In: Cam, L.L. and Neyman, J., Eds., Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, The University of California Press, 281-297.
[19] 孙翠羽, 董倩, 王益文, 等. 基于机器学习的翻车事故严重程度分析方法[J]. 交通工程, 2024, 24(11): 68-77.
[20] 李子辉. 基于数据驱动的高速公路交通安全风险识别与预测研究[D]: [硕士学位论文]. 济南: 山东建筑大学, 2023.
[21] 董春娇, 许博, 李鹏辉, 等. 轨迹数据驱动的高速公路事故风险研判[J]. 北京交通大学学报, 2024, 48(6): 12-21.