Please wait a minute...
Journal of Zhejiang University (Science Edition)  2024, Vol. 51 Issue (2): 143-152    DOI: 10.3785/j.issn.1008-9497.2024.02.002
Geographic Information Science     
A road traffic accident risk assessment method considering the arrival time cost
Keran SUN1(),Yingzhi WANG2,Feng ZHANG1,3(),Renyi LIU1,3
1.School of Earth Sciences,Zhejiang University,Hangzhou 310058,China
2.Department of Traffic Management Engineering,Zhejiang Police College,Hangzhou 310053,China
3.Zhejiang Provincial Key Laboratory of Geographic Information Science,Zhejiang University,Hangzhou 310058,China
Download: HTML( 2 )   PDF(4146KB)
Export: BibTeX | EndNote (RIS)      

Abstract  

Road traffic accidents occur frequently and have significant impacts on life, property, and society. However, existing researches on road traffic accident risk pay little attention on establishing an effective road network model that accurately describes the transmission characteristics of traffic accident risk. As a result, the accuracy of risk evaluation is limited. To address this issue, we propose a network geographically weighted regression method based on arrival time cost. We conduct experiments using data from roads, traffic violations, traffic accidents, and urban points of interest (POIs) in a city from 2018 to 2020. The experimental results demonstrate that the network geographically weighted regression method, based on arrival time cost, incorporates the propagation nature of traffic accident risk on the road. It significantly reduces evaluation errors and effectively evaluates road traffic accident risk and its influencing factors. The downtown area of the city exhibits high accident risk, primarily concentrated at intersections with heavy traffic flow and certain road points with inadequate transportation facilities. The impact of different types of road traffic violations and urban POIs on the risk of road traffic accidents varies significantly and exhibits strong spatial heterogeneity.



Key wordsroad traffic accidents      cost network geographically weighted regression      arrival time cost      spatial analysis     
Received: 01 August 2022      Published: 08 March 2024
CLC:  P 208  
Corresponding Authors: Feng ZHANG     E-mail: krsun@zju.edu.cn;zfcarnation@zju.edu.cn
Cite this article:

Keran SUN,Yingzhi WANG,Feng ZHANG,Renyi LIU. A road traffic accident risk assessment method considering the arrival time cost. Journal of Zhejiang University (Science Edition), 2024, 51(2): 143-152.

URL:

https://www.zjujournals.com/sci/EN/Y2024/V51/I2/143


考虑抵达时间成本的道路交通事故风险评估方法

道路交通事故频发,给生命财产造成重大损失,给社会生活带来重大影响。现有针对道路交通事故风险的研究未建立有效的道路网络模型,难以准确描述交通事故风险在道路上的传播特点,评估准确度不高。基于此,提出了一种基于抵达时间成本的网络地理加权回归方法,并利用某县级市2018—2020年的道路、交通违法、交通事故、城市POI等数据开展实验,结果表明,基于抵达时间成本的网络地理加权回归方法融合了交通事故风险在道路上的传播性质,显著降低了评估误差,能够有效评估道路交通事故风险及其影响因素;市中心区域道路交通事故高风险区域主要集中在车流量较大的道路交会处与部分交通设施尚不完备的道路;各类交通违法数量、城市POI对道路交通事故风险的影响程度不同,且具有很强的空间异质性。


关键词: 道路交通事故,  成本网络地理加权回归,  抵达时间成本,  空间分析 
Fig.1 Map of the administrative scope and main roads of the research area

道路交通事故

风险类别

250 m道路分段当量

死亡人数之和/人

一类>5.5
二类(3.5,5.5]
三类(1.5,3.5]
Table 1 Road traffic accident risk category
Fig.2 The spread of road traffic accident risk on the road
网络数据集字段类型说明算法
成本近似通行时间成本Ct
限制单行限制B/F
Table 2 Road travel time cost network dataset parameters
道路类型平均单向车道数平均限速/(km·h-1
高速公路4100
国道(穿越城市)350
省道(穿越城市)350
县道(穿越城市)250
乡镇村道120
市区一级道路460
市区二级道路340
其他道路225
Table 3 Average number of lanes and average speed limits for different types of roads

变量

类型

英文标识符号说明
因变量DeY当量死亡人数
自变量battleX1争抢类违法数量
drinksX2影响驾驶行为违法数量
reverseX3逆向行驶违法数量
overspeedX4超速行驶违法数量
signalX5违反交通信号违法数量
carX61 km内汽车服务设施数量
entertainmentX71 km内娱乐服务设施数量
foodX81 km内餐饮服务设施数量
trafficX91 km内交通设施数量
Table 4 Regression model variable
Fig.3 Flow chart of road traffic accident risk assessment based on arrival time CNGWR model
自变量VIF
X11.007
X21.218
X31.565
X41.009
X51.522
X62.065
X714.393
X812.294
X94.038
Table 5 Multicollinearity test results
指标数值
赤池信息准则1.599 7
最佳带宽(适应型)3
Table 6 Best bandwidth and Akaike information criterion results
回归模型

预测值平均

相对误差/%

全局线性回归模型155.32
欧氏距离地理加权回归模型67.35
基于抵达时间的CNGWR模型16.94
Table 7 Average relative error of different regression models for road traffic accident risk assessment
Fig.4 Distribution of road traffic accident risk in the research area
Fig.5 Distribution of road traffic accident high risk point in the center of the research area
Fig.6 The regression coefficient of each influencing factor of road traffic accident risk in the center of the research area
Fig.7 Regression coefficient of influencing factors of some high-risk accident points in the center of the research area
[16]   黄晓丽, 刘耀龙, 段锦, 等. 基于灰色关联及模糊综合评价法的道路交通安全风险评价[J]. 数学的实践与认识, 2017, 47(7): 208-215.
HUANG X L, LIU Y L, DUAN J, et al. The assessment of road traffic safety risk based on grey relative and fuzzy comprehensive evaluation method[J]. Journal of Mathematics in Practice and Theory, 2017, 47(7): 208-215.
[17]   蔡晓禹, 雷财林, 彭博, 等. 基于驾驶行为和信息熵的道路交通安全风险预估[J]. 中国公路学报, 2020, 33(6): 190-201. DOI:10.3969/j.issn.1001-7372. 2020.06.018
CAI X Y, LEI C L, PENG B, et al. Road traffic safety risk estimation based on driving behavior and information entropy[J]. China Journal of Highway and Transport, 2020, 33(6): 190-201. DOI:10. 3969/j.issn.1001-7372.2020.06.018
doi: 10. 3969/j.issn.1001-7372.2020.06.018
[18]   樊亚新. 城市道路交通事故时空分析与多约束空间分区优化研究[D]. 武汉: 武汉大学, 2019. DOI:10.27379/d.cnki.gwhdu.2019.000169
FAN Y X. Research on Network Spatiotemporal Analysis and Multi-constraint Spatial Regionalization for Urban Traffic Collisions[D]. Wuhan: Wuhan University, 2019. DOI:10.27379/d.cnki.gwhdu.2019. 000169
doi: 10.27379/d.cnki.gwhdu.2019. 000169
[19]   孟庆达, 于龙, 尹皓邈, 等. 基于GIS的道路交通事故空间分布特征分析:以山东省日照市为例[J]. 信息与电脑(理论版), 2021, 33(20): 50-53. DOI:10.3969/j.issn.1003-9767.2021.20.016
MENG Q D, YU L, YIN H M, et al. Analysis of spatial distribution characteristics of road traffic accidents based on GIS: Taking Rizhao city, Shandong province as an example[J]. China Computer & Communication, 2021, 33(20): 50-53. DOI:10.3969/j.issn.1003-9767.2021.20.016
doi: 10.3969/j.issn.1003-9767.2021.20.016
[1]   国家统计局. 中国统计年鉴[M]. 北京: 中国统计出版社, 2021: 781.
National Bureau of Statistics. China Statistical Yearbook[M]. Beijing: China Statistics Press, 2021: 781.
[2]   王长君, 高岩, 张爱红. 重点违法行为导致交通事故的数据分析[J]. 交通运输工程与信息学报, 2005(3): 29-36. DOI:10.3969/j.issn.1672-4747.2005. 03.004
WANG C J, GAO Y, ZHANG A H. Analysis of the road traffic fatalities caused by severe traffic violation[J]. Journal of Transportation Engineering and Information, 2005(3): 29-36. DOI:10.3969/j.issn. 1672-4747.2005.03.004
doi: 10.3969/j.issn. 1672-4747.2005.03.004
[3]   刘林才, 付川云. 交通违法行为与事故关系研究:基于零截尾负二项和广义泊松模型[J]. 综合运输, 2021, 43(12): 51-58.
LIU L C, FU C Y. Relationship between traffic violations and crashes based on zero-truncated negative binomial and generalized Poisson models[J]. China Transportation Review, 2021, 43(12): 51-58.
[4]   朱彤, 秦丹, 董傲然, 等. 公交驾驶员违规类型同交通事故间的关联分析[J]. 交通运输系统工程与信息, 2022, 22(2): 322-329. DOI:10.16097/j.cnki. 1009-6744.2022.02.033
ZHU T, QIN D, DONG A R, et al. Relational analysis between bus drivers' violation type and traffic accident[J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(2): 322-329. DOI:10.16097/j.cnki.1009-6744. 2022.02.033
doi: 10.16097/j.cnki.1009-6744. 2022.02.033
[5]   郭淼, 赵晓华, 姚莹, 等. 基于驾驶行为和交通运行状态的事故风险研究[J].华南理工大学学报(自然科学版), 2022, 50(9): 29-38. DOI:10.12141/j.issn. 1000-565X.210629
GUO M, ZHAO X H, YAO Y, et al. Study on accident risk based on driving behavior and traffic operating status[J]. Journal of South China University of Technology (Natural Science Edition), 2022, 50(9): 29-38. DOI:10.12141/j.issn.1000-565X.210629
doi: 10.12141/j.issn.1000-565X.210629
[6]   王颖志, 沈雅婕, 王立君. 基于改进兴趣度度量与Apriori算法的交通事故多发点成因分析[J]. 浙江大学学报(理学版), 2021, 48(3): 349-355. DOI:10.3785/j.issn.1008-9497.2021.03.011
WANG Y Z, SHEN Y J, WANG L J. The causes analysis of traffic accident black spots based on improved interest measurement and Apriori algorithm[J]. Journal of Zhejiang University (Science Edition), 2021, 48(3): 349-355. DOI:10.3785/j.issn.1008-9497.2021. 03.011
doi: 10.3785/j.issn.1008-9497.2021. 03.011
[7]   高红丽, 邓昌俊, 王文迪, 等. 基于国家车辆事故深度调查体系的道路交通事故驾驶员人因分析[J]. 科学技术与工程, 2019, 19(19): 342-348.
GAO H L, DENG C J, WANG W D, et al. Analysis of driver's cause of road traffic accident based on the national automobile accident in-depth investigation system[J]. Science Technology and Engineering, 2019, 19(19): 342-348.
[8]   谢学斌, 孔令燕. 基于ARIMA和XGBoost组合模型的交通事故预测[J]. 安全与环境学报, 2021, 21(1): 277-284. DOI:10.13637/j.issn.1009-6094. 2019.0633
XIE X B, KONG L Y. On the ways to the traffic accident prediction based on the ARIMA and XGBoost combined model[J]. Journal of Safety and Environment, 2021, 21(1): 277-284. DOI:10.13637/j.issn.1009-6094.2019.0633
doi: 10.13637/j.issn.1009-6094.2019.0633
[9]   陈玉飞, 魏思怡, 张林. 基于GM(1,N)的道路交通事故预测模型[J]. 华北理工大学学报(自然科学版), 2020, 42(1): 47-50. DOI:10.3969/j.issn.2095-2716.2020.01.009
CHEN Y F, WEI S Y, ZHANG L. Forecast model for road traffic accidents based on GM(1,N) model[J]. Journal of North China University of Science and Technology (Natural Science Edition), 2020, 42(1): 47-50. DOI:10.3969/j.issn.2095-2716.2020.01.009
doi: 10.3969/j.issn.2095-2716.2020.01.009
[10]   陈碧瑶. 基于Bi-LSTM的道路交通事故数预测研究[D]. 西安: 长安大学, 2021. DOI:10.26976/d.cnki.gchau.2021.001367
CHEN B Y. Research on Forecast of Road Traffic Accidents Based on Bi-LSTM[D]. Xi'an: Chang'an University, 2021. DOI:10.26976/d.cnki.gchau. 2021.001367
doi: 10.26976/d.cnki.gchau. 2021.001367
[11]   张光南, 钟俏婷, 杨清玄. 交通违法事故时空分布特征及其影响因素:以广州市为例[J]. 交通运输系统工程与信息, 2019, 19(3): 208-214. DOI:10.16097/j.cnki.1009-6744.2019.03.031
ZHANG G N, ZHONG Q T, YANG Q X. Temporal-spatial characteristics and influencing factors of at-fault traffic crashes: A case study in Guangzhou[J]. Journal of Transportation Systems Engineering and Information Technology, 2019, 19(3): 208-214. DOI:10.16097/j.cnki.1009-6744. 2019.03.031
doi: 10.16097/j.cnki.1009-6744. 2019.03.031
[12]   LEVINE N, KIM K E, NITZ L H. Spatial analysis of Honolulu motor vehicle crashes: II. Zonal generators[J]. Accident Analysis & Prevention, 1995, 27(5): 675-685. DOI:10.1016/0001-4575(95)00018-u
doi: 10.1016/0001-4575(95)00018-u
[13]   FLAHAUT B. Impact of infrastructure and local environment on road unsafety: Logistic modeling with spatial autocorrelation[J]. Accident Analysis & Prevention, 2004, 36(6): 1055-1066. DOI:10.1016/j.aap.2003.12.003
doi: 10.1016/j.aap.2003.12.003
[14]   ERDOGAN S. Explorative spatial analysis of traffic accident statistics and road mortality among the provinces of Turkey[J]. Journal of Safety Research, 2009, 40(5): 341-351. DOI:10.1016/j.jsr. 2009. 07.006
doi: 10.1016/j.jsr. 2009. 07.006
[1] CHEN Wanlong, FENG Youjian. Research on the regional network structure of the Yangtze River Delta from the perspective of railway passenger transport[J]. Journal of Zhejiang University (Science Edition), 2021, 48(5): 606-616.
[2] LU Lu, SUN Gennian. Crisis cycle of inbound tourism in the Chinese mainland from 2008 to 2018 and its market attribution[J]. Journal of Zhejiang University (Science Edition), 2021, 48(3): 377-390.
[3] WU Xiaojun, FANG Xiuqin, MIAO Yuexian, WANG Kai, PANG Chanyun. Study on urban expansion of Hefei city based on Landsat data[J]. Journal of Zhejiang University (Science Edition), 2017, 44(6): 631-639.
[4] ZHAO Taotao, BAI Jianjun, SHANG Zhonghui. Research on rurality differentiation of county areas in Shaanxi Province based on BP neural network[J]. Journal of Zhejiang University (Science Edition), 2016, 43(2): 203-210.