Quantifying the Weather Impact on Urban Transport Reliability using XGBoost and SHAP

DOI: https://doi.org/10.33445/sds.2025.15.6.23
Keywords: public transport reliability; machine learning; explainable AI; risk management; meteorological factors; gradient boosting; critical infrastructure

Abstract

Purpose. The purpose of the work is to develop and verify a method for quantitatively assessing the impact of a complex of meteorological factors on operational failures of public transport.

Method. The study is based on machine learning methods. The high-precision ensemble gradient boosting algorithm XGBoost was used to predict delays. For an in-depth analysis of risk factors and overcoming the «black box» problem, an advanced interpreted machine learning tool SHAP was used, which can establish the contribution of each factor to a specific forecast.

Findings. It was established that meteorological conditions are a significant, nonlinear risk factor. The analysis showed that the greatest threat to traffic stability is posed by temperature anomalies (a sharp decrease), significant precipitation and high wind speed. Using the proposed method, it was proven that the impact of weather is heterogeneous, i.e. insignificant under normal conditions and a strong trigger of large-scale failures during rush hours.

Practical implications. The results of the study create a scientific basis for the development of information systems to support decision-making in the field of crisis management. The proposed method makes it possible to move from responding to changes that have already occurred to implementing preventive measures: informing passengers in advance, optimizing resources and increasing the resilience of the transport system to climate threats.

Paper type. Empirical.

Downloads

Download data is not yet available.

Author Biographies

Yurii Matseliukh, Lviv Polytechnic National University

Yurii Matseliukh received the Bachelor’s degree in Computer Science (with honours) and the Master’s degree in System Analysis (with honours) from Lviv Polytechnic National University, Lviv, Ukraine.

He is currently a PhD student in the Information Systems and Networks Department at Lviv Polytechnic National University, Lviv, Ukraine. For the past decade, he has served on the review board for scientific journals in the field of transportation and transport systems, such as Simulation Modelling Practice and Theory. He is the author of more than 30 publications (a Scopus h-index =13). His research focuses on modelling passenger transportation in public transport.

MSс Matseliukh was a recipient of the Presidential Scholarship of Ukraine from 2019 to 2021. His research has been presented at numerous international conferences, including the IEEE International Scientific and Technical Conference on Computer Sciences and Information Technologies.

Vasyl Lytvyn, Lviv Polytechnic National University

Vasyl Lytvyn holds the Doctor of Technical Sciences degree and Professor of the Information Systems and Networks Department at the Lviv Polytechnic National University, Lviv, Ukraine. He graduated from Ivan Franko National University of Lviv, Ukraine, in 1997.
He has been a Head of the Information Systems and Networks Department at Lviv Polytechnic National University, Lviv, Ukraine (2013-2025). Prof. Lytvyn V. is an expert in Data Science, Big Data, Intelligent Systems, Machine Learning, Knowledge Engineering and Ontology Construction. His main direction of scientific research is the development of intelligent decision support systems using an ontological approach. He is participated in much research as a performer and supervisor. He is the author of more than 200 publications in journals including the International Journal of Computing, Symmetry, Applied System Innovation et al. Prof. Lytvyn‘s publications have a Hirsch index of 31 in the Scopus database.

References

Komarnytsky, I.M., Bublyk, M.I. (2008). Assessment of man-made damages and analysis of approaches to their calculation in global and regional aspects. Bulletin of Lviv Polytechnic National University. Series: Economics and Management Issues, 628, 134–144.

Matseliukh Y., & Lytvyn, V. (2025). Application of the K-means clustering method in the organisation of passenger transport in a smart city. Innovative Technologies and Scientific Solutions for Industries, 1 (31), 83–101. https://doi.org/10.30837/2522-9818.2025.1.083

Matseliukh Y., & Lytvyn, V. (2025). Concept of a digital twin of passenger transportation by public transport in a smart city. Таuridа Scientific Herald. Series: Technical Sciences, 1(4), 425-436. https://doi.org/10.32782/tnv-tech.2025.4.1.40

Pasichnyk, A., Pasichnyk, V., & Kushchenko, Y. (2021). Application of system approach in methodology of solving engineering problems. Transport Systems and Transportation Technologies, (21), 23. https://doi.org/10.15802/tstt2021/237635

Rachynska, A.V. Classification of risks of the railways as a basis for formation of economic security for its functioning. Economy and Society, 6, 81-87.

Royko Yu., Bura R., Maksymiuk S. Determination of delays on two-lane streets within public transport stops. Advances In Mechanical Engineering and Transport, 2(11), 24-28.

Alonso-Fernandez, F., Butt, T. H., and Tiwari, P. (2025). Overtake detection in trucks using CAN bus signals: A comparative study of machine learning methods. Expert Systems with Applications, 297, 129339. https://doi.org/10.1016/j.eswa.2025.129339

Arunachalam, S. (2025). Long-term and short-term traffic flow prediction with different weather conditions using optimized wheel-graph attention based neural network. Future Generation Computer Systems, 176, 108198. https://doi.org/10.1016/j.future.2025.108198

Banach, M., and Długosz, R. (2025). Solutions for planning smart hybrid public transportation system based on Google Maps and Voronoi diagrams. Journal of Computational and Applied Mathematics, 472, 116775. https://doi.org/10.1016/j.cam.2025.116775

Chukhray, N., Shakhovska, N., Mrykhina, O., et al (2019). Consumer aspects in assessing the suitability of technologies for the transfer. In 2019 IEEE 14th International Conference on Computer Sciences and Information Technologies (CSIT) (pp. 142–147). https://doi.org/10.1109/stc-csit.2019.8929879

Explain XGBoost Predictions with SHAP. (2025). XGBoosting.com. Retrieved from https://xgboosting.com/explain-xgboost-predictions-with-shap/

Gao, P., Shuai, B., Zhang, R., and Wang, B. (2025). STCAGNN-RNKDE: a traffic accident prediction model for spatiotemporal combinatorial attention graph neural networks using Ripley’s K and network kernel density estimation. Reliability Engineering and System Safety, 265, 111593. https://doi.org/10.1016/j.ress.2025.111593

Gong, Y., Yu, P., Zhang, X., Zhang, X., Nie, X., and Sun, H. (2025). Exploiting dynamic spatio-temporal correlations for origin-destination demand prediction. Expert Systems With Applications, 299, 130095. https://doi.org/10.1016/j.eswa.2025.130095

Gu, S., Miao, A., Yan, D., Hu, B., Cheng, G., and Xu, Z. (2025). Resilience Assessment of Urban Conventional Bus Networks in Cold Regions under Snowfall Weather. Reliability Engineering and System Safety. https://doi.org/10.1016/j.ress.2025.111824

Guo, Z., Zhou, X., Lu, C., Li, S., Zhu, Y., Liu, Y., and Li, Z. (2025). Urban agglomeration transportation resilience: evaluation and evolution analysis using a data-driven model. Environmental and Sustainability Indicators, 26, 100714. https://doi.org/10.1016/j.indic.2025.100714

Ha, T. V., and Arimura, M. (2025). Mapping urban mobility during extreme weather events using mobile spatial statistics data. Transportation Research Interdisciplinary Perspectives, 33, 101609. https://doi.org/10.1016/j.trip.2025.101609

Hanbarikarekani, M., Qu, X., Zeibots, M., and Qi, W. (2018). Minimizing the average delay at intersections via presignals and speed control. Journal of Advanced Transportation, 2018, 4121582. https://doi.org/10.1155/2018/4121582

Kaynak, S., Kaynak, B., Mermer, O., and Demir, I. (2025). City-scale digital twin framework for flood impact analysis: Integrating urban infrastructure and real-time data analytics. Urban Climate, 64, 102640. https://doi.org/10.1016/j.uclim.2025.102640

Kuzmin, O., and Bublyk, M. (2016). Economic evaluation and government regulation of technogenic (man-made) damage in the national economy. In 2016 XIth International Scientific and Technical Conference Computer Sciences and Information Technologies (CSIT).

Liu, L., Wang, R., Tian, T., and Li, S. (2025). Hazardous materials transportation route optimization considering dynamic wind speeds and risk equity. Journal of Loss Prevention in the Process Industries, 98, 105748. https://doi.org/10.1016/j.jlp.2025.105748

Ma, F., Zhang, Y., Liu, Q., Guo, Y., Yang, Z., and Zhang, J. (2025). Modeling urban transportation safety resilience under extreme rainstorms: A catastrophe theory approach. Reliability Engineering and System Safety, 263, 111301. https://doi.org/10.1016/j.ress.2025.111301

Maksoud, N., AlJassmi, H., Ali, L., and Masoud, A. R. (2025). Applications of large language models and generative AI in transportation: A systematic review and bibliometric analysis. Transportation Research Interdisciplinary Perspectives, 34, 101699. https://doi.org/10.1016/j.trip.2025.101699

Masmoudi, M., Shakrouf, Y., Omar, O. H., Shikhli, A., Abdalla, F., Alketbi, W., ... and Siam, A. I. (2025). Driver risk classification for transportation safety: A machine learning approach using psychological, physiological, and demographic factors with driving simulator. Engineering Applications of Artificial Intelligence, 162, 112585. https://doi.org/10.1016/j.engappai.2025.112585

Parsa, A. B., Movahedi, A., Taghipour, H., Derrible, S., and Mohammadian, A. (K.). (2020). Toward safer highways, application of XGBoost and SHAP for real-time accident detection and feature analysis. Accident Analysis and Prevention, 136, 105405. https://doi.org/10.1016/j.aap.2019.105405

Saki, S., and Soori, M. (2025). Artificial intelligence, machine learning and deep learning in advanced transportation systems, a review. Multimodal Transportation, 5, 100242. https://doi.org/10.1016/j.multra.2025.100242

Theivadas, J. R., and Ponnan, D. S. (2026). DrivNet: Adaptive driving behavior prediction for improving the safety of multimodal transportation in autonomous vehicle systems. Multimodal Transportation, 5, 100266. https://doi.org/10.1016/j.multra.2025.100266

Wang, X., Pan, Z., Liu, P., Wang, X., Fan, Z., Liu, N., and Wu, J. (2025). The interaction of weather factors on multimodal public transportation: An empirical research on the combination of bike-sharing and metro. Cities, 169, 106537. https://doi.org/10.1016/j.cities.2025.106537

Zhang, C., Zou, X., and Lin, C. (2022). Fusing XGBoost and SHAP Models for Maritime Accident Prediction and Causality Interpretability Analysis. Journal of Marine Science and Engineering, 10(8), 1154. https://doi.org/10.3390/jmse10081154

Zhang, X., Gao, R., Xiao, Z., Wang, K., Liu, T., Liang, M., and Zhang, J. (2025а). Natural language processing and text mining in transportation: Current status, challenges, and future roadmap. Expert Systems with Applications, 296, 129050. https://doi.org/10.1016/j.eswa.2025.129050

Zhang, Z., Yang, J., Chen, X., and Wu, Y. (2025b). Graph neural networks with flow conservation constraints for real-time origin-destination matrix completion. Pattern Recognition, 170, 112046. https://doi.org/10.1016/j.patcog.2025.112046

Zhao, N., Zhang, Y., Lu, Y., and Zhai, G. (2025). Spatial heterogeneity and multimodal complementarity in travel behavior resilience under extreme rainfall: A machine learning approach. Journal of Transport Geography, 130, 104472. https://doi.org/10.1016/j.jtrangeo.2025.104472

Zhou, L., Wang, Z., Mao, Z., Wang, Q., Zhang, D., Zhang, Z., and Chen, Z. (2025). Data-driven remaining driving range estimation and analysis framework for electric vehicles under real-world conditions. Applied Energy, 402, 126881. https://doi.org/10.1016/j.apenergy.2025.126881


Abstract views: 73
PDF Downloads: 30
Published
2025-12-31
How to Cite
Matseliukh, Y., & Lytvyn, V. (2025). Quantifying the Weather Impact on Urban Transport Reliability using XGBoost and SHAP. Social Development and Security, 15(6), 260-274. https://doi.org/10.33445/sds.2025.15.6.23
Issue
Vol. 15 No. 6 (2025)
Section
Engineering and Technology

Copyright (c) 2025 Yurii Matseliukh, Vasyl Lytvyn

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

The authors agree with the following conditions:

1. Authors retain copyright and grant the journal right of first publication (Download agreement) with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.

2. Authors have the right to complete individual additional agreements for the non-exclusive spreading of the journal’s published version of the work (for example, to post work in the electronic repository of the institution or to publish it as part of a monograph), with the reference to the first publication of the work in this journal.

3. Journal’s politics allows and encourages the placement on the Internet (for example, in the repositories of institutions, personal websites, SSRN, ResearchGate, MPRA, SSOAR, etc.) manuscript of the work by the authors, before and during the process of viewing it by this journal, because it can lead to a productive research discussion and positively affect the efficiency and dynamics of citing the published work (see The Effect of Open Access).