A Spatiotemporal Machine Learning Framework for Predicting Supply Chain Disruptions Under Extreme Weather Conditions

References

[1] Y. Bhattarai, V. Chaudhary, C. Walker, R. Talchabhadel, and S. Sharma, “Ensemble learning for enhancing critical infrastructure resilience to urban flooding,” Scientific Reports , vol. 15, no. 1, p. 36901, Oct. 2025, 10.1038/s41598-025-20970-2 .

[2] J. Li et al. , “Prediction of Power Failure Caused by Extreme Weather Based on Spatiotemporal Data,” pp. 777– 780, Nov. 2024, 10.1109/icnepe64067.2024.10860621 .

[3] S. M. A. B. Dong, A. M. A. X. Li, and A. M. A. L. Wu, “A GNN-Based Framework for Assessing Flood Impacts on Highway Networks: Integrating Network Structural, Functional, and Social Features,” Journal of Management in Engineering , vol. 42, no. 4, Mar. 2026, 10.1061/jmenea.meeng-7198 .

[4] F. Yuan et al. , “Predicting road flooding risk with crowdsourced reports and fine-grained traffic data,” Computational Urban Science , vol. 3, no. 1, Mar. 2023,

[5] X. Yang, “O 2 RDL-net for joint risk classification and delay forecasting in logistics systems using interaction amplified deep,” Scientific Reports , May 2026, 10.1038/s41598-026-55703-6 .

[6] X. Chen, “Design of supply chain risk prediction and early warning model based on deep neural network,” pp. 1053– 1058, Jul. 2025,

[7] S. O. Usoro, I. A. Omogiate, R. Z. Felix, and D. A. Galadima, “AI-Driven Risk Forecasting for Strengthening the United States Food Supply Chain Resilience: Case Study: A National AI-Enabled Food Supply Chain Risk Forecasting Framework and a Data Analytics Approach to Predicting Disruptions,” International Journal of Science and Research Archive , vol. 18, no. 1, pp. 744–756, Jan. 2026,

[8] D. Effah, C. Bai, W. A. Asante, and M. Quayson, “The Role of Artificial Intelligence in Coping With Extreme Weather- Induced Cocoa Supply Chain Risks,” IEEE Transactions on Engineering Management , vol. 71, pp. 9854–9875, Jul. 2023,

[9] D. Huang, S. Wang, and Z. Liu, “A systematic review of prediction methods for emergency management,” International Journal of Disaster Risk Reduction , vol. 62, p. 102412, Jun. 2021,

[10] C. Fan, X. Jiang, and A. Mostafavi, “A network percolation-based contagion model of flood propagation and recession in urban road networks,” Scientific Reports , vol. 10, no. 1, p. 13481, Aug. 2020, 020-70524-x .

[11] D. J. Morrice, P. Cronin, F. Tanrısever, and J. C. Butler, “Supporting hurricane inventory management decisions with consumer demand estimates,” Journal of Operations Management , vol. 45, no. 1, pp. 86–100, Jul. 2016, 10.1016/j.jom.2016.05.006 .

[12] R. Kumar, M. Bhanu, J. Mendes‐Moreira, and J. Chandra, “Spatio-Temporal Predictive Modeling Techniques for Different Domains: a Survey,” ACM Computing Surveys , vol. 57, no. 2, pp. 1–42, Sep. 2024,

[13] D. Ivanov, “Simulation-based ripple effect modelling in the supply chain,” International Journal of Production Research , vol. 55, no. 7, pp. 2083–2101, Jan. 2017, 10.1080/00207543.2016.1275873 .

[14] A. Dolgui, D. Ivanov, and B. Sokolov, “Ripple effect in the supply chain: an analysis and recent literature,” International Journal of Production Research , vol. 56, pp. 414–430, Oct. 2017, 10.1080/00207543.2017.1387680 .

[15] A. Kinra, D. Ivanov, A. Das, and A. Dolgui, “Ripple effect quantification by supplier risk exposure assessment,” International Journal of Production Research , vol. 58, no. 18, pp. 5559 –5578, Oct. 2019, 10.1080/00207543.2019.1675919 .

[16] W. J. Tan, W. Cai, and A. N. Zhang, “Structural-aware simulation analysis of supply chain resilience,” International Journal of Production Research , vol. 58, no. 17, pp. 5175 –5195, Dec. 2019, 10.1080/00207543.2019.1705421 .

[17] K. Benidis et al. , “Deep Learning for Time Series Forecasting: Tutorial and Literature Survey,” arXiv (Cornell University) , vol. 55, no. 6, pp. 1–36, May 2022,

[18] I. Vallés-Ṕerez, E. Soria‐Olivas, M. Martínez‐Sober, A. J. Serrano-López, J. Gómez‐Sanchís, and F. Mateo, “Approaching sales forecasting using recurrent neural networks and transformers,” Expert Systems with Applications , vol. 201, p. 116993, Apr. 2022, 10.1016/j.eswa.2022.116993 .

[19] G. Theodoridis and A. Tsadiras, “Retail Demand Forecasting: A Comparative Analysis of Deep Neural Networks and the Proposal of LSTMixer, a Linear Model Extension,” Information , vol. 16, no. 7, p. 596, Jul. 2025,

[20] N. S. Sukanya, V. V. Asha, M. T. Vasumathi, S. Gothe, S. Fernandes, and S. Kumar, “Enhancing Weather Prediction for Transportation and Logistics Using the Temporal Fusion Transformer Algorithm,” pp. 1161–1165, Jun. 2025,

[21] I. Zeroual and J. E. Bouhdidi, “Enhancing supply chain agility with advanced weather forecasting,” International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering , vol. 15, no. 6, p. 5904, Nov. 2025, 10.11591/ijece.v15i6.pp5904-5913 .

[22] A. Alzahrani and M. Z. Asghar, “Intelligent Risk Prediction System in IoT-Based Supply Chain Management in Logistics Sector,” Electronics , vol. 12, no. 13, p. 2760, Jun. 2023,

[23] B. Abdulla and B. Birgisson, “Predicting Road Network Vulnerability to Fluvial Flooding Using Machine Learning Classifiers: Case Study of Houston during Hurricane Harvey,” Construction Research Congress 2020 , pp. 38– 47, Nov. 2020,

[24] S. Takyi, R. B. Antwi, E. E. Özgüven, L. Okine, and R. Moses, “Towards Sustainable and Resilient Infrastructure: Hurricane-Induced Roadway Closure and Accessibility Assessment in Florida Using Machine Learning,” Sustainability , vol. 17, no. 9, p. 3909, Apr. 2025, 10.3390/su17093909 .

[25] H. Yang, D. Duque, and D. P. Morton, “Optimizing diesel fuel supply chain operations to mitigate power outages for hurricane relief,” IISE Transactions , vol. 54, no. 10, pp. 936–949, Mar. 2022, 10.1080/24725854.2021.2021461 .

[26] M. Hoque, A. R. Nabil, K. Hossain, N. Akhtar, and M. R. Amin, “Integrated AI-Based Decision Support Systems for Emergency Supply Chain Management in the United States During Natural Disasters,” Journal of Computer Science and Technology Studies , vol. 7, no. 10, pp. 275–289, Oct. 2025,

[27] W. A. Zogaan, N. Ajabnoor, and A. A. Salamai, “Leveraging deep learning for risk prediction and resilience in supply chains: insights from critical industries,” Journal Of Big Data , vol. 12, no. 1, Apr. 2025, 10.1186/s40537-025-01143-4 .

[28] Z. Shen, H. Wang, J. Chen, and X. Song, “Resilience Inference for Supply Chains with Hypergraph Neural Network,” in Proceedings of the AAAI Conference on Artificial Intelligence , Association for the Advancement of Artificial Intelligence, Mar. 2026, pp. 39259–39267. 10.1609/aaai.v40i46.41274 .

[29] Sichong Huang, “AI-Driven Early Warning Systems for Supply Chain Risk Detection: A Machine Learning Approach,” Academic Journal of Computing & Information Science , vol. 8, no. 9, Jan. 2025, 10.25236/ajcis.2025.080913 .

[30] D. Ivanov and A. Dolgui, “Stress testing supply chains and creating viable ecosystems,” Operations Management Research , vol. 15, pp. 475–486, May 2021, 10.1007/s12063-021-00194-z .

[31] Md. I. Hossain, S. Talapatra, P. Saha, and H. M. Belal, “From Theory to Practice: Leveraging Digital Twin Technologies and Supply Chain Disruption Mitigation Strategies for Enhanced Supply Chain Resilience with Strategic Fit in Focus,” Global Journal of Flexible Systems Management , vol. 26, no. 1, pp. 87–109, Dec. 2024, 10.1007/s40171-024-00424-w .

[32] N. K. Jingar, “Ensuring Safety, Accountability, and Drift Resistance in LLM-Based Supply Chain Optimization,” International Journal of Scientific Research in Science Engineering and Technology , p. 472, Jan. 2023, 10.32628/ijsrset2310372 .

[33] S. Petrova and M. Hughes1, “Physics-Informed Graph Neural Networks for Supply Chain Disruption Prediction and Mitigation,” Frontiers in Applied Physics and Mathematics , vol. 2, no. 1, pp. 130–147, Dec. 2025, 10.71465/fapm458 .

[34] Z. Ge, L. Kong, and A. Brintrup, “Federated machine learning for privacy preserving, collective supply chain risk prediction,” International Journal of Production Research , vol. 61, no. 23, pp. 8115–8132, Feb. 2023, 10.1080/00207543.2022.2164628 .

@article{1725048,
      author = {Sohail Sayed, Nauman Sayed, Muhammad Khurram Khan},
      title = {A Spatiotemporal Machine Learning Framework for Predicting Supply Chain Disruptions Under Extreme Weather Conditions},
      journal = {Iconic Research And Engineering Journals},
      year = {2025},
      volume = {9},
      number = {2},
      pages = {1614-1622},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1725048.pdf},
      abstract = {Extreme weather is among the most disruptive forces acting on modern supply chains: globally, more than two million kilometers of urban road networks are exposed to 1-in-100-year flood events, and in the United States alone extreme weather caused over 1.2 million vehicle crashes, 5,000 road deaths, and 400,000 injuries annually between 2007 and 2016 [1]. Yet most supply-chain risk models treat weather as a static covariate rather than a spatiotemporal process that propagates through transportation and distribution networks. This paper proposes a spatiotemporal machine learning framework — ST-EWS — that couples (i) a spatiotemporal weather encoder based on convolutional-LSTM residual networks and temporal fusion transformers, (ii) a graph-neural-network network-impact model that translates hazard fields into link-serviceability and node-disruption probabilities over the logistics graph, (iii) a demand-shock forecaster for essential goods, and (iv) a multi-level alerting layer with quantified lead time and explainability. The framework consolidates the reported evidence envelope: ConvLSTM-based spatiotemporal models reach 90.46% prediction accuracy for weather-induced infrastructure faults [2]; graph attention hazard models predict flood impact on highway networks at better accuracy than structural-feature baselines and identify the most vulnerable corridors before and during events [3]; tree-based road-inundation models reach AUC 0.860 for Hurricane Harvey and 0.790 for Tropical Storm Imelda [4]; hybrid deep networks classify logistics risk at 92% accuracy with SHAP-validated interpretability [5]; and spatiotemporal attention networks achieve 92.7% accuracy for emergency-level supply-chain alerts with an 8.3-second response delay and a 7.3% false-alarm rate [6]. The paper argues that weather resilience in essential-goods supply chains is fundamentally a spatiotemporal prediction problem, and that closing it requires fusing meteorology, network topology, and demand dynamics in one trainable loop.},
      keywords = {essential goods, extreme weather, flood risk, graph neural networks, Spatiotemporal machine learning, supply chain disruption prediction, supply chain resilience, transportation networks},
      month = {August},
  }