Abstract
Highway flood-damage blocking poses a critical threat to transportation system resilience, yet risk assessment is often limited by insufficient modeling of temporal disaster evolution, highly imbalanced data, and weak model interpretability. To address these challenges, this study proposes an integrated modeling framework that combines temporal data augmentation, machine learning, and interpretable mechanism analysis. Three data-balancing strategies—Time-series Generative Adversarial Network (TimeGAN) augmentation, undersampling, and a hybrid approach—were systematically compared to handle imbalanced temporal data. Six machine learning models (Logistic Regression, Support Vector Machine, Random Forest, Decision Tree, eXtreme Gradient Boosting, and Multilayer Perceptron) were evaluated, and SHapley Additive exPlanations (SHAP) was used to quantify factor contributions and explore nonlinear effects and interaction patterns. Results show that the Multilayer Perceptron trained on TimeGAN-augmented sequences achieved the highest performance, with an F1 score of 49.81% and PR-AUC of 49.46%. SHAP analysis identified key drivers and their threshold effects: Daily precipitation exceeding 2.8 mm, 7-day effective precipitation (EP 7) exceeding 22 mm, temperature above 21 °C, and average road-stream distance within 1 km (ARSD) above 0.15 km significantly increase the risk of highway flood-damage blocking. High temperature conditions are more likely to coincide with heavy precipitation and elevated EP 7, and their combined effects further amplify blocking risk. Factor contributions also varied across methods, reflecting SHAP’s ability to capture nonlinear effects and reveal interaction patterns, whereas linear regression mainly reflects independent linear effects. By integrating temporal generation, systematic model evaluation, and interpretable analysis, this study enhances the accuracy and reliability of highway flood-damage blocking prediction, providing quantitative guidance for flood damage prevention and resilience improvement of highway systems.
Similar content being viewed by others
Data availability
The public data that support the findings of this study are available on request from the corresponding author.
References
Zheng, J. et al. Construction of an intelligent risk identification system for highway flood damage based on multimodal large models. Appl. Sci. 15, 12782 (2025).
Du, B. et al. Assessing the impact of precipitation variability on landslide hazards in urbanized regions. Int. J. Appl. Earth Obs. Geoinf. 136, 104360 (2025).
Xin, Z. et al. The relationship between geological disasters with land use change, meteorological and hydrological factors: A case study of Neijiang City in Sichuan Province. Ecol. Ind. 154, 110840 (2023).
Liu, D. et al. A new method for calculating highway blocking due to high-impact weather conditions. Nat. Hazard. 25, 493–513 (2025).
Pan, J. et al. Characterizing China’s road network development from a spatial entropy perspective. J. Transp. Geogr. 116, 103848 (2024).
Ministry of Transport of the People’s Republic of China. Statistical Communique on the Development of the Transportation Industry (2024). https://xxgk.mot.gov.cn/2020/jigou/zhghs/202506/t20250610_4170228.html (2025).
Zhang, W. et al. A year marked by extreme precipitation and floods: Weather and climate extremes in 2024. Adv. Atmos. Sci. 42, 1045–1063 (2025).
Smyl, D., Ghasemzadeh, F. & Pour-Ghaz, M. Modeling water absorption in concrete and mortar with distributed damage. Constr. Build. Mater. 125, 438–449 (2016).
Kringos, N. & Scarpas, A. Raveling of asphaltic mixes due to water damage. Transp. Res. Record: J. Transp. Res. Board 1929, 79–87 (2005).
Zeng, Y., Wang, Q., Cao, J. & Liu, G. (2020) Study on Water Damage Mechanism and Emergency Restore of Fill Subgrade upon Squashy Slope Foundation in Mountain Area. IOP Conf. Ser.: Earth Environ. Sci. 455: 012181.
Sun, D., Wen, H., Zhang, Y. & Xue, M. An optimal sample selection-based logistic regression model of slope physical resistance against rainfall-induced landslide. Nat Hazards 105, 1255–1279 (2021).
Mahmoud, A. A. et al. Synergizing machine learning and experimental analysis to predict post-heating compressive strength in waste concrete. Struct. Concr. 26, 2916–2950 (2025).
Li, Z. et al. Spatiotemporal assessment of water damage susceptibility in China’s road infrastructure: a machine learning and SHAP approach using social media data. J. Hydrol. 664, 134539 (2026).
Wang, L. et al. Time series prediction of reservoir bank landslide failure probability considering the spatial variability of soil properties. J. Rock Mech. Geotech. Eng. 16, 3951–3960 (2024).
Bashir, N. et al. Enhancing seismic activity classification in augmented soil gas radon time series data through computational intelligence techniques. J. Atmos. Solar Terr. Phys. 274, 106560 (2025).
EskandariNasab, M., Hamdi, S. M. & Filali Boubrahimi, S. Impacts of data preprocessing and sampling techniques on solar flare prediction from multivariate time series data of photospheric magnetic field parameters. ApJS 275, 6 (2024).
Wang, K., Yang, T., Kong, S. & Li, M. Air quality index prediction through TimeGAN data recovery and PSO-optimized VMD-deep learning framework. Appl. Soft Comput. 170, 112626 (2025).
Ali, R., Muayad, M., Mohammed, A. S. & Asteris, P. G. Analysis and prediction of the effect of Nanosilica on the compressive strength of concrete with different mix proportions and specimen sizes using various numerical approaches. Struct. Concr. 24, 4161–4184 (2022).
Zeyad, A. M. et al. Compressive strength of nano concrete materials under elevated temperatures using machine learning. Sci. Rep. 14, 24246 (2024).
Mahmoud, A. A., El-Sayed, A. A., Aboraya, A. M., Fathy, N. & Nabil, I. M. Enhancing predictive accuracy of nano-additive concrete gamma ray attenuation at high temperatures using AI-based models. Neural Comput. Applic. 37, 21833–21866 (2025).
Liu, Y. et al. Regional sustainable development strategy based on the coordination between ecology and economy: A case study of Sichuan Province China. Ecol. Indicators 134, 108445 (2022).
Department of Transportation of Sichuan Province. Statistical Bulletin on the Development of the Transportation Sector in Sichuan Province (2023). https://jtt.sc.gov.cn/jtt/c101520/2024/10/15/5f5e139f8e8c4c4397440e0805e04bd9.shtml (2024).
Li, H. et al. A daily gap-free normalized difference vegetation index dataset from 1981 to 2023 in China. Sci Data 11, 527 (2024).
Bandara, C. M. M. Drainage density and effective precipitation. J. Hydrol. 21, 187–190 (1974).
China Meteorological Administration. Meteorological risk early warning levels of geological disaster induced by torrential rain. Standard QX/T 487–2019 (2019).QX/T 487—2019. (2019).
Yan, Y. et al. Spatial Distribution-based Imbalanced Undersampling. IEEE Trans. Knowl. Data Eng. 35(6), 6376–6391. https://doi.org/10.1109/tkde.2022.3161537 (2022).
Chen, C., Wang, F., Wang, Z., Zhang, D. & Xiang, L. A novel flood forecasting model based on TimeGAN for data-sparse basins. Stoch. Env. Res. Risk Assess. 39, 2267–2280 (2025).
Zhang, C. & Ji, D. HAL-Net: A historical analogy learning network for adaptive and interpretable pandemic forecasting. Expert Syst. Appl. 299, 130038 (2026).
Wang, L., Xiao, M., Lv, J. & Liu, J. Analysis of influencing factors of traffic accidents on urban ring road based on the SVM model optimized by Bayesian method. PLoS ONE 19, e0310044 (2024).
Bai, J. et al. Multinomial random forest. Pattern Recogn. 122, 108331 (2022).
He, Z., Wu, Z., Xu, G., Liu, Y. & Zou, Q. Decision Tree for Sequences. IEEE Trans. Knowl. Data Eng. 35(1), 251–263. https://doi.org/10.1109/tkde.2021.3075023 (2021).
Yan, X. et al. Driving risk prediction of urban arterial and collector roads using multi-dimensional real-time data. Eng. Appl. Artif. Intell. 138, 109386 (2024).
Betkier, I. Estimating travel time in transport network with a combined multi-attributed graph convolutional neural network and multilayer perceptron model. Eng. Appl. Artif. Intell. 142, 109898 (2025).
Wang, X., Jin, Y., Schmitt, S. & Olhofer, M. Recent Advances in Bayesian Optimization. ACM Comput. Surv. 55, 1–36 (2023).
Fujiwara, K. Knowledge distillation with resampling for imbalanced data classification: Enhancing predictive performance and explainability stability. Results Eng. 24, 103406 (2024).
López-Oriona, Á. & Vilar, J. A. The bootstrap for testing the equality of two multivariate time series with an application to financial markets. Inf. Sci. 616, 255–275 (2022).
Fathy, I. N., Dahish, H. A., Alkharisi, M. K., Mahmoud, A. A. & Fouad, H. E. E. Predicting the compressive strength of concrete incorporating waste powders exposed to elevated temperatures utilizing machine learning. Sci. Rep. 15, 25275 (2025).
Pradhan, S., Toll, D. G., Rosser, N. J. & Brain, M. J. An investigation of the combined effect of rainfall and road cut on landsliding. Eng. Geol. 307, 106787 (2022).
Ashland, F. X. Critical shallow and deep hydrologic conditions associated with widespread landslides during a series of storms between February and April 2018 in Pittsburgh and vicinity, western Pennsylvania, USA. Landslides https://doi.org/10.1007/s10346-021-01665-x (2021)
Sun, W., Li, J., Yu, R., Li, N. & Zhang, Y. Exploring changes of precipitation extremes under climate change through global variable-resolution modeling. Science Bulletin 69, 237–247 (2024).
Ran, Q. et al. The relative importance of antecedent soil moisture and precipitation in flood generation in the middle and lower Yangtze River basin. Hydrol. Earth Syst. Sci. 26, 4919–4931 (2022).
Ye, S. et al. From rainfall to runoff: The role of soil moisture in a mountainous catchment. J. Hydrol. 625, 130060 (2023).
Zhang, L., Zhang, X., Gao, S. & Gu, X. Revealing nonlinear relationships and thresholds of human activities and climate change on ecosystem services in Anhui Province based on the XGBoost–SHAP model. Sustainability 17, 8728 (2025).
Funding
This research was generously funded by the National Key R&D Program of China (Grant No. 2024YFB4303100) on Integrated Technology Application for Autonomous Transportation Systems.
Author information
Authors and Affiliations
Contributions
Bin Li: Conceptualization, Validation, Supervision, Funding acquisition, Writing—review & editing; Lingyi Wu: Writing—review & editing, Writing—original draft, Methodology, Formal analysis, Data curation; Jian Gao: Resources, Project administration; Feng Yang: Resources, Project administration; Yuqi Guo: Validation, Investigation; Mingyue Yan: Writing—review & editing.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Li, B., Wu, L., Gao, J. et al. Comparative analysis of machine learning models with SHAP interpretation for causes of highway flood-damage blocking. Sci Rep (2026). https://doi.org/10.1038/s41598-026-35074-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-35074-8
