Forecasting the Severity of Road Accidents in Russian Regions Using Interpretable Machine Learning Methods

Daniil V. Sobolev, Angi E. Skhvediani, Maxim I. Gerashchenko, Tatiana Yu. Kudryavtseva

Peter the Great St. Petersburg Polytechnic University, Saint-Petersburg, Russia

Abstract

Predicting the severity of road traffic accidents is an important applied task for reducing mortality and injury rates. Given the high dimensionality, heterogeneity, and class imbalance of data, traditional statistical models do not always provide adequate predictive performance and are limited in capturing complex non-linear relationships. This study aims to develop and comparatively analyze machine learning models for predicting crash severity using micro-level data from the Northwestern Federal District of Russia, as well as to interpret the contribution of key factors to the final prediction. The study tests two hypotheses: first, behavioral and spatial-infrastructure characteristics contribute more to predicting severe outcomes than meteorological and lighting conditions; second, ensemble machine learning methods provide higher predictive performance compared to linear and basic tree-based algorithms. The empirical base comprises data on 167,000 road traffic accidents that occurred between 2015 and 2024. The research procedure involved aggregating hierarchical records into an analytical table, generating the feature space, balancing classes, building and comparing models, and interpreting the results using SHAP analysis. The evaluated models included logistic regression, decision trees, gradient boosting ensembles, and fully connected neural networks with categorical embeddings. The experimental results showed that an ensemble of gradient boosting and a neural network demonstrated the best predictive performance (ROC AUC = 0.746). It was found that geographic location, vehicle characteristics, and the behavioral patterns of road users make the greatest contribution to predicting severe outcomes, whereas the contribution of weather conditions is secondary. The obtained results support both research hypotheses. The theoretical significance of this study lies in advancing the predictive approach to traffic accident analysis using explainable machine learning. The practical significance is driven by the potential to use the findings for developing intelligent transportation systems (ITS) and prioritizing road safety improvement measures.

Keywords

accident prediction; machine learning; gradient boosting; neural networks; model interpretability; SHAP; road safety; risk factor analysis.

JEL classification

References

1. Ahmed, S.K., Mohammed, M.G., Abdulqadir, S.O., El-Kader, R.G.A., El‐Shall, N.A., Chandran, D., Rehman, M.E.U., Dhama, K. (2023). Road traffic accidental injuries and deaths: A neglected global health issue. Health Science Reports, Vol. 6, Issue 5, e1240. https://doi.org/10.1002/hsr2.1240

          2. Verma, R., Agarwal, M.M. (2026). A systematic review on road accident prediction: A special attention toward machine learning and deep learning approaches. Journal of Circuits, Systems and Computers, Vol. 35, No. 2, 2530009. https://doi.org/10.1142/S0218126625300090

          3. Zhang, J., Li, Z., Pu, Z., Xu, C. (2018). Comparing prediction performance for crash injury severity among various machine learning and statistical methods. IEEE Access, Vol. 6, 60079–60087. https://doi.org/10.1109/ACCESS.2018.2874979

          4. Skhvediani, A., Rodionova, M., Savchenko, N., Kudryavtseva, T. (2023). Prediction of the road accidents severity level: Case of Saint-Petersburg and Leningrad Oblast. International Journal of Technology, Vol. 14, No. 8, 1717–1727. https://doi.org/10.14716/ijtech.v14i8.6859

          5. Rodionova, M., Skhvediani, A., Kudryavtseva, T. (2021). Determinants of pedestrian–vehicle crash severity: Case of Saint Petersburg, Russia. International Journal of Technology, Vol. 12, No. 7, 1427–1436. https://doi.org/10.14716/ijtech.v12i7.5403

          6. Rodionova, M., Skhvediani, A., Kudryavtseva, T. (2022). Prediction of crash severity as a way of road safety improvement: The case of Saint Petersburg, Russia. Sustainability, Vol. 14, Issue 16, 9840. https://doi.org/10.3390/su14169840

          7. Jamal, A., Zahid, M., Tauhidur Rahman, M., Al-Ahmadi, H.M., Almoshaogeh, M., Farooq, D., Ahmad, M. (2021). Injury severity prediction of traffic crashes with ensemble machine learning techniques: A comparative study. International Journal of Injury Control and Safety Promotion, Vol. 28, Issue 4, 408–427. https://doi.org/10.1080/17457300.2021.1928233

          8. Mohammadpour, S.I., Khedmati, M., Zada, M.J.H. (2023). Classification of truck-involved crash severity: Dealing with missing, imbalanced, and high dimensional safety data. PLoS ONE, Vol. 18, Issue 3, e0281901. https://doi.org/10.1371/journal.pone.0281901

          9. Najafi Moghaddam Gilani, V., Hosseinian, S.M., Ghasedi, M., Nikookar, M. (2021). Data-driven urban traffic accident analysis and prediction using logit and machine learning-based pattern recognition models. Mathematical Problems in Engineering, Vol. 2021, 1–11. https://doi.org/10.1155/2021/9974219

          10. Tang, J., Liang, J., Han, C., Li, Z., Huang, H. (2019). Crash injury severity analysis using a two-layer Stacking framework. Accident Analysis & Prevention, Vol. 122, 226–238. https://doi.org/10.1016/j.aap.2018.10.016

          11. Boo, Y., Choi, Y. (2022). Comparison of mortality prediction models for road traffic accidents: An ensemble technique for imbalanced data. BMC Public Health, Vol. 22, Issue 1, 1476. https://doi.org/10.1186/s12889-022-13719-3

          12. Azhar, A., Ariff, N.M., Bakar, M.A.A., Roslan, A. (2022). Classification of driver injury severity for accidents involving heavy vehicles with Decision Tree and Random Forest. Sustainability, Vol. 14, Issue 7, 4101. https://doi.org/10.3390/su14074101

          13. Sanjurjo-de-No, A., Pérez-Zuriaga, A.M., García, A. (2023). Analysis and prediction of injury severity in single micromobility crashes with Random Forest. Heliyon, Vol. 9, Issue 12, e23062. https://doi.org/10.1016/j.heliyon.2023.e23062

          14. Santos, D., Saias, J., Quaresma, P., Nogueira, V.B. (2021). Machine learning approaches to traffic accident analysis and hotspot prediction. Computers, Vol. 10, Issue 12, 157. https://doi.org/10.3390/computers10120157

          15. Komol, M.M.R., Hasan, M.M., Elhenawy, M., Yasmin, S., Masoud, M., Rakotonirainy, A. (2021). Crash severity analysis of vulnerable road users using machine learning. PLoS ONE, Vol. 16, Issue 8, e0255828. https://doi.org/10.1371/journal.pone.0255828

          16. Das, S., Tamakloe, R., Zubaidi, H., Obaid, I., Ashifur Rahman, M.B. (2023). Bicyclist injury severity classification using a random parameter logit model. International Journal of Transportation Science and Technology, Vol. 12, Issue 4, 1093–1108. https://doi.org/10.1016/j.ijtst.2023.02.001

          17. Siregar, M.L., Tjahjono, T., Yusuf, N. (2022). Predicting the segment-based effects of heterogeneous traffic and road geometric features on fatal accidents. International Journal of Technology, Vol. 13, No. 1, 92–102. https://doi.org/10.14716/ijtech.v13i1.4450

          18. Chang, F.-R., Huang, H.-L., Schwebel, D.C., Chan, A.H.S., Hu, G.-Q. (2020). Global road traffic injury statistics: Challenges, mechanisms and solutions. Chinese Journal of Traumatology, Vol. 23, Issue 4, 216–218. https://doi.org/10.1016/j.cjtee.2020.06.001

          19. Mehrannia, P., Bagi, S.S.G., Moshiri, B., Al-Basir, O.A. (2023). Deep representation of imbalanced spatio‐temporal traffic flow data for traffic accident detection. IET Intelligent Transport Systems, Vol. 17, Issue 3, 606–619. https://doi.org/10.1049/itr2.12287

          20. Li, P., Abdel-Aty, M., Yuan, J. (2020). Real-time crash risk prediction on arterials based on LSTM-CNN. Accident Analysis & Prevention, Vol. 135, 105371. https://doi.org/10.1016/j.aap.2019.105371

          21. Calsavara, F., Kabbach, F.I., Larocca, A.P.C. (2021). Effects of fog in a Brazilian road segment analyzed by a driving simulator for sustainable transport: Drivers’ speed profile under in-vehicle warning systems. Sustainability, Vol. 13, Issue 19, 10501. https://doi.org/10.3390/su131910501

          22. Calsavara, F., Kabbach Junior, F.I., Larocca, A.P.C. (2021). Effects of fog in a Brazilian road segment analyzed by a driving simulator for sustainable transport: Drivers’ visual profile. Sustainability, Vol. 13, Issue 16, 9448. https://doi.org/10.3390/su13169448

          23. Karim, F., Albrka Ali, S.I. (2020). Evaluation of most influential factors affecting road traffic accidents in Sidon, Lebanon. Jurnal Kejuruteraan, Vol. 32, Issue 4, 645–650. https://doi.org/10.17576/jkukm-2020-32(4)-12

          24. Zuraida, R., Abbas, B.S. (2020). The factors influencing fatigue related to the accident of intercity bus drivers in Indonesia. International Journal of Technology, Vol. 11, No. 2, 342–352. https://doi.org/10.14716/ijtech.v11i2.3792

          25. Khurshid, A., Sohail, A., Khurshid, M., Shah, M.U., Jaffry, A.A. (2021). Analysis of road traffic accident fatalities in Karachi, Pakistan: An autopsy-based study. Cureus, Vol. 13, Issue 4, e14459. https://doi.org/10.7759/cureus.14459

          26. Bohn, B., Garcke, J., Iza-Teran, R., Paprotny, A., Peherstorfer, B., Schepsmeier, U., Thole, C.-A. (2013). Analysis of car crash simulation data with nonlinear machine learning methods. Procedia Computer Science, Vol. 18, 621–630. https://doi.org/10.1016/j.procs.2013.05.226

          27. Anderson, T.K. (2009). Kernel density estimation and K-means clustering to profile road accident hotspots. Accident Analysis & Prevention, Vol. 41, Issue 3, 359–364. https://doi.org/10.1016/j.aap.2008.12.014

          28. Kang, Y., Khattak, A.J. (2022). Deep learning model for crash injury severity analysis using Shapley Additive Explanation values. Transportation Research Record: Journal of the Transportation Research Board, Vol. 2676, Issue 12, 242–254. https://doi.org/10.1177/03611981221095087

          29. Benfaress, I., Bouhoute, A., Zinedine, A. (2024). Enhancing traffic accident severity prediction using ResNet and SHAP for interpretability. AI, Vol. 5, Issue 4, 2568–2585. https://doi.org/10.3390/ai5040124

          30. Le, K.G. (2025). Improving road safety: Supervised machine learning analysis of factors influencing crash severity. Scientific Journal of Silesian University of Technology. Series Transport, Vol. 127, 129–153. https://doi.org/10.20858/sjsutst.2025.127.8

          31. Fiorentini, N., Losa, M. (2020). Handling imbalanced data in road crash severity prediction by machine learning algorithms. Infrastructures, Vol. 5, Issue 7, 61. https://doi.org/10.3390/infrastructures5070061

          32. Obasi, I.C., Benson, C. (2023). Evaluating the effectiveness of machine learning techniques in forecasting the severity of traffic accidents. Heliyon, Vol. 9, Issue 8, e18812. https://doi.org/10.1016/j.heliyon.2023.e18812

          33. Alogaili, A., Mannering, F. (2022). Differences between day and night pedestrian-injury severities: Accounting for temporal and unobserved effects in prediction. Analytic Methods in Accident Research, Vol. 33, 100201. https://doi.org/10.1016/j.amar.2021.100201

          34. Chen, S., Shao, H., Ji, X. (2021). Insights into factors affecting traffic accident severity of novice and experienced drivers: A machine learning approach. International Journal of Environmental Research and Public Health, Vol. 18, Issue 23, 12725. https://doi.org/10.3390/ijerph182312725

          35. Zhang, G., Xuan, Q., Cai, Y., Hu, X., Hu, N., Jiang, X., Yao, X. (2025). Identifying the reciprocal causation between hit-and-run behavior and crash injury severity. Traffic Injury Prevention, Vol. 26, Issue 2, 156–163. https://doi.org/10.1080/15389588.2024.2402464

          36. Roussou, S., Ziakopoulos, A., Yannis, G. (2025). Investigation of hit-and-run crash severity through explainable machine learning. Transportation Letters, Vol. 17, Issue 10, 1755–1770. https://doi.org/10.1080/19427867.2025.2488976

          37. Parenteau, C.S., Viano, D.C. (2014). Light-vehicle occupancy and severe injury by vehicle and crash type. Traffic Injury Prevention, Vol. 15, Issue 5, 457–461. https://doi.org/10.1080/15389588.2013.835044

          38. Ahmed, S., Hossain, M.A., Ray, S.K., Bhuiyan, M.M.I., Sabuj, S.R. (2023). A study on road accident prediction and contributing factors using explainable machine learning models: Analysis and performance. Transportation Research Interdisciplinary Perspectives, Vol. 19, 100814. https://doi.org/10.1016/j.trip.2023.100814

Acknowledgements

This research was funded by the Russian Science Foundation (project No. 23-78-10176, https://rscf.ru/en/project/23-78-10176/)

About Authors

Daniil Vasilievich Sobolev

Research Assistant, Research Laboratory «System Dynamics», Higher School of Engineering and Economics, Institute of Industrial Management of Economics and Trade, Peter the Great St. Petersburg Polytechnic University, Saint-Petersburg, Russia (195251, Saint-Petersburg, Polytechnicheskaya street, 29); ORCID: https://orcid.org/0009-0000-5869-5317 e-mail: sobolevdan2002@gmail.com

Angi Erastievich Skhvediani

Candidate of Economic Sciences, Head of Scientific Research Laboratory «System Dynamics», Associate Professor, Higher School of Engineering and Economics, Institute of Industrial Management of Economics and Trade, Peter the Great St. Petersburg Polytechnic University, Saint-Petersburg, Russia (195251, Saint-Petersburg, Polytechnicheskaya street, 29); ORCID https://orcid.org/0000-0001-7171-7357 e-mail: shvediani_ae@spbstu.ru

Maxim Ivanovich Gerashchenko

Research Assistant, Research Laboratory «System Dynamics», Higher School of Engineering and Economics, Institute of Industrial Management of Economics and Trade, Peter the Great St. Petersburg Polytechnic University, Saint-Petersburg, Russia (195251, Saint-Petersburg, Polytechnicheskaya street, 29); ORCID: https://orcid.org/0009-0004-0383-9204 e-mail: Rosefourx@gmail.com

Tatiana Yurievna Kudryavtseva

Doctor of Economics, Professor, Higher School of Engineering and Economics, Institute of Industrial Management of Economics and Trade, Peter the Great St. Petersburg Polytechnic University Saint-Petersburg, Russia (195251, Saint-Petersburg, Polytechnicheskaya street, 29); ORCID: https://orcid.org/0000-0003-1403-3447 e-mail: kudryavtseva_tyu@spbstu.ru

For citation

Sobolev, D.V., Skhvediani, A.E., Gerashchenko, M.I., Kudryavtseva, T.Yu. (2026). Forecasting the Severity of Road Accidents in Russian Regions Using Interpretable Machine Learning Methods. Journal of Applied Economic Research, Vol. 25, No. 2, 596-624. https://doi.org/10.15826/vestnik.2026.25.2.020

Article info

Received February 16, 2026; Revised March 25, 2026; Accepted March 28, 2026.

DOI: http://dx.doi.org/10.15826/vestnik.2026.25.2.020

Download full text article: