Modeling and Predicting Urban Traffic Accident Severity via Intelligent Algorithms

Document Type : Original Article

Authors
Alireza Taherian 1
Saeid Sherafatipour 2
Amirhossein Ghiyami 1
Mahmoud Saffarzadeh 3
1 M.Sc., Grad., Civil and Environmental Engineering, Department of Transportation Planing and Engineering, Tarbiat Modarres University, Tehran, Iran.
2 Assistant Professor, Civil and Environmental Engineering, Department of Transportation Planing and Engineering, Tarbiat Modarres University, Tehran, Iran.
3 Professor, Civil and Environmental Engineering, Department of Transportation Planing and Engineering, Tarbiat Modarres University, Tehran, Iran.
10.22034/road.2025.524307.2401
Abstract
Urban traffic accidents represent a significant challenge in Iran, primarily due to inadequate infrastructure, lack of adherence to traffic regulations, and the absence of a well-established driving culture. These factors have led to considerable human and economic losses. In contrast, many developed countries have achieved substantial reductions in urban traffic accidents through the strategic application of technology, public education, and intelligent urban planning. The success of these nations highlights the critical role of effective traffic management in enhancing urban safety.This study aims to identify and predict the severity of urban road accidents in Alborz Province using a decision tree algorithm. For this purpose, a dataset comprising 5,000 recorded accidents from 2018 to 2022 was collected and, after data cleansing, analyzed in the Python programming environment. The results revealed that the three most influential variables in predicting accident severity are the type of vehicle collision, the time of occurrence, and whether the accident took place on a non-working day. The final model achieved a prediction accuracy of 0.79, indicating a relatively high level of precision in distinguishing between injury and fatal accidents. The findings suggest that, in addition to human-related factors, environmental and structural elements such as the nature of vehicle collisions significantly contribute to the likelihood of fatal outcomes. Leveraging real-world data and an interpretable modeling approach, this research offers practical insights for road safety policymaking and the design of effective preventive strategies.
Keywords
Accident Severity
Decision Tree
Urban Accidents
Traffic Safety
-Abrari Vajari, Mohammad, Kayvan Aghabayk, Mohammad Sadeghian, and Nirajan Shiwakoti (2020). A Multinomial Logit Model of Motorcycle Crash Severity at Australian Intersections. Journal of Safety Research 73 (June):17–24.
doi.org/10.1016/j.jsr.2020.02.008
-Aron, Maurice, Romain Billot, Nour-Eddin El Faouzi, and Régine Seidowsky (2015). Traffic Indicators, Accidents and Rain: Some Relationships Calibrated on a French Urban Motorway Network. Transportation Research Procedia 10:31–40.
doi.org/10.1016/j.trpro.2015.09.053
-Colagrande, Sandro (2022). A Methodology for the Characterization of Urban Road Safety through Accident Data Analysis.” Transportation Research Procedia 60:504–11. doi.org/10.1016/j.trpro.2021.12.065
-Eliane, Hernandez Garcia Michelle, and Lozano Angelica. (2024). Analysis of Road Accidents in Two Mixed Industrial Urban Zones, Using Nested Poisson and Negative Binomial Models. Transportation Research Procedia 78:377–83. doi.org/10.1016/j.trpro.2024.02.048
-EslamiNezhad, S. A., and M. R. Delavar (2019). An Integrated Network-Constrained Spatial Analysis For Car Accidents: A Case Study Of Tehran City, Iran. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLII-4/W18 (October):335–42.
doi.org/10.5194/isprs-archives-XLII-4-W18-335-2019
-García, María Flor, Armando Ortuño Padilla, Begoña Guirao Abad, and Jairo Casares Blanco. (2021). Urban Road Accidents and Ride-Hailing Services: A Study of Dependence in Madrid. Transportation Research Procedia 58:301–8. doi.org/10.1016/j.trpro.2021.11.041
-Hyodo, Satoshi, and Tomoyuki Todoroki (2018). An Analysis of Risk Factors for
Rear-Ender Accident on Urban Expressway Considering Accident Severity. Transportation Research Procedia 34:203–10. doi.org/10.1016/j.trpro.2018.11.033
-Kaplan, Sigal, and Carlo Giacomo Prato (2012). Risk Factors Associated with Bus Accident Severity in the United States: A Generalized Ordered Logit Model. Journal of Safety Research 43 (3): 171–80.
doi.org/10.1016/j.jsr.2012.05.003
-Rezapour, Mahdi, Ahmed Farid, Sahima Nazneen, and Khaled Ksaibati (2021). Using Machine Leaning Techniques for Evaluation of Motorcycle Injury Severity. IATSS Research 45 (3): 277–85.
doi.org/10.1016/j.iatssr.2020.07.004
-Santos, Kenny, João P. Dias, and Conceição Amado (2022). A Literature Review of Machine Learning Algorithms for Crash Injury Severity Prediction. Journal of Safety Research 80 (February), 254–69.
doi.org/10.1016/j.jsr.2021.12.007
-Wang, Deyu, Qinyi Liu, Liang Ma, Yijing Zhang, and Haozhe Cong (2019). Road Traffic Accident Severity Analysis: A Census-Based Study in China. Journal of Safety Research 70 (September):135–47. doi.org/10.1016/j.jsr.2019.06.002
-Yoon, Junho (2025). Prediction of High-Risk Areas Using the Interpretable Machine Learning: Based on Each Determinant for the Severity of Pedestrian Crashes. Journal of Transport Geography 126 (June):104216.
doi.org/10.1016/j.jtrangeo.2025.104216
-Yuan, Yalong, Min Yang, Yanyong Guo, Soora Rasouli, Zuoxian Gan, and Yifeng Ren (2021). Risk Factors Associated with Truck-Involved Fatal Crash Severity: Analyzing Their Impact for Different Groups of Truck Drivers. Journal of Safety Research 76 (February):154–65. doi.org/10.1016/j.jsr.2020.12.012