Investigating Factors Influencing Crash Severity on Mountainous Two-Lane Roads: Machine Learning Versus Statistical Models

被引:2
作者
Qi, Ziyuan [1 ]
Yao, Jingmeng [1 ]
Zou, Xuan [1 ]
Pu, Kairui [1 ]
Qin, Wenwen [1 ,2 ]
Li, Wu [1 ,2 ]
机构
[1] Kunming Univ Sci & Technol, Fac Transportat Engn, Kunming 650500, Peoples R China
[2] Kunming Univ Sci & Technol, Yunnan Integrated Transport Dev & Reg Logist Manag, Kunming 650500, Peoples R China
关键词
traffic safety; mountainous two-lane roads; machine learning; statistical model; SHAP; sustainable transportation; DRIVER INJURY SEVERITY; ORDERED PROBIT; TRAFFIC ACCIDENTS; MULTINOMIAL LOGIT; RISK-FACTORS;
D O I
10.3390/su16187903
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Due to poor road design, challenging terrain, and difficult geological conditions, traffic accidents on mountainous two-lane roads are more frequent and severe. This study aims to address the lack of understanding of key factors affecting accident severity with the goal of improving mountainous traffic safety, thereby contributing to sustainable transportation systems. The focus of this study is to compare the interpretability of model performances with three statistical models (Ordered Logit, Partial Proportional Odds Model, and Multinomial Logit) and six machine learning models (Decision Tree, Random Forest, Gradient Boosting, Extra Trees, AdaBoost, and XGBoost) on two-lane mountain roads in Yunnan Province, China. Additionally, we assessed the ability of these models to uncover underlying causal relationships, particularly how accident causes affect severity. Using the SHapley Additive exPlanations (SHAP) method, we interpreted the influence of risk factors in the machine learning models. Our findings indicate that machine learning models, especially XGBoost, outperform statistical models in predicting accident severity. The results highlight that accident patterns are the most significant determinants of severity, followed by road-related factors and the type of colliding vehicles. Environmental factors like weather, however, have minimal impact. Notably, vehicle falling, head-on collisions, and longitudinal slope sections are linked to more severe accidents, while minor accidents are more frequent on horizontal curve sections and areas that combine curves and slopes. These insights can help traffic management agencies develop targeted measures to reduce accident rates and enhance road safety, which is critical for promoting sustainable transportation in mountainous regions.
引用
收藏
页数:27
相关论文
共 59 条
  • [11] Predicting effects of built environment on fatal pedestrian accidents at location-specific level: Application of XGBoost and SHAP
    Chang, Iljoon
    Park, Hoontae
    Hong, Eungi
    Lee, Jaeduk
    Kwon, Namju
    [J]. ACCIDENT ANALYSIS AND PREVENTION, 2022, 166
  • [12] Differences in injury severity of accidents on mountainous highways and non-mountainous highways
    Chen, Feng
    Chen, Suren
    [J]. INTELLIGENT AND INTEGRATED SUSTAINABLE MULTIMODAL TRANSPORTATION SYSTEMS PROCEEDINGS FROM THE 13TH COTA INTERNATIONAL CONFERENCE OF TRANSPORTATION PROFESSIONALS (CICTP2013), 2013, 96 : 1868 - 1879
  • [13] Accident factor analysis based on different age groups via AdaBoost algorithm
    Chen, Lihui
    Wang, Pin
    [J]. CANADIAN JOURNAL OF CIVIL ENGINEERING, 2019, 46 (05) : 364 - 370
  • [14] Severity Analysis of Multi-Truck Crashes on Mountain Freeways Using a Mixed Logit Model
    Chen, Zheng
    Wen, Huiying
    Zhu, Qiang
    Zhao, Sheng
    [J]. SUSTAINABILITY, 2023, 15 (08)
  • [15] Identifying significant predictors of injury severity in traffic accidents using a series of artificial neural networks
    Delen, D
    Sharda, R
    Bessonov, M
    [J]. ACCIDENT ANALYSIS AND PREVENTION, 2006, 38 (03) : 434 - 444
  • [16] Modeling and comparing injury severity of at-fault and not at-fault drivers in crashes
    Duddu, Venkata R.
    Penmetsa, Praveena
    Pulugurtha, Srinivas S.
    [J]. ACCIDENT ANALYSIS AND PREVENTION, 2018, 120 : 55 - 63
  • [17] A proactive decision support system for predicting traffic crash events: A critical analysis of imbalanced class distribution
    Elamrani Abou Elassad, Zouhair
    Mousannif, Hajar
    Al Moatassime, Hassan
    [J]. KNOWLEDGE-BASED SYSTEMS, 2020, 205
  • [18] Fernández-Delgado M, 2014, J MACH LEARN RES, V15, P3133
  • [19] Analysis of accident injury-severity outcomes: The zero-inflated hierarchical ordered probit model with correlated disturbances
    Fountas, Grigorios
    Anastasopoulos, Panagiotis Ch
    [J]. ANALYTIC METHODS IN ACCIDENT RESEARCH, 2018, 20 : 30 - 45
  • [20] Greedy function approximation: A gradient boosting machine
    Friedman, JH
    [J]. ANNALS OF STATISTICS, 2001, 29 (05) : 1189 - 1232