A Data-Driven Two-Stage Robust Optimization Model for Sponge Iron Supply Chains under Market Volatility and Dual-Season Energy Disruption Risks

Mohammad Reza  Badami; Roya M. Ahari; Mahsa  Ghandehari; Mahtab  Sherafati

Authors

Mohammad Reza Badami Department of Industrial Engineering, Na.C., Islamic Azad University, Najafabad, Iran Author https://orcid.org/0009-0000-0036-5665
Roya M. Ahari Modern Manufacturing Research Center, Na.C., Islamic Azad University, Najafabad, Iran Corresponding author https://orcid.org/0000-0002-8952-8682
Mahsa Ghandehari Department of Management, Faculty of Administrative Sciences and Economics, University of Isfahan, Isfahan, Iran Author https://orcid.org/0000-0003-1864-6881
Mahtab Sherafati Department of Industrial Engineering, Na.C., Islamic Azad University, Najafabad, Iran Author https://orcid.org/0009-0008-6967-9310

Keywords:

Data-driven robust optimization, Sponge iron supply chain, Energy disruptions, Inventory management , Deep learning residuals, Exact mathematical programming

Abstract

Energy-intensive process industries, particularly direct-reduced iron (DRI) and sponge iron manufacturing, face critical operational vulnerabilities stemming from dual-season energy disruptions—such as winter natural gas curtailments and summer electricity blackouts—compounded by volatile raw material market prices. Traditional deterministic planning and decoupled forecasting models often fail to account for overlapping systemic shocks, frequently leading to costly production halts or severe inventory stockouts. To bridge this gap, this study proposes a novel two-stage data-driven robust optimization model for sponge iron supply chains. The framework systematically translates empirical forecasting error residuals derived from deep learning models into polyhedral uncertainty sets governed by budgets of uncertainty ( ). Validated through a real-world case study of Mobarakeh Steel Company, the numerical results demonstrate that setting optimal uncertainty budgets at and achieves an optimal compromise point, yielding a 14.2% total operational cost reduction compared to deterministic baselines while completely preventing stockout-induced production halts.

References

[1] L. O'Connor, "A Comparative Review of Material Supply Chains and Sustainability Pathways in Steel and Battery Technologies," The Extractive Industries and Society, vol. 46, p. 101560, 2026, doi: 10.1016/j.exis.2025.101814.

[2] F. Wu, J. Gao, Y. Tong, H. Fang, G. Li, and T. Yue, "Carbon Footprint Characteristics and Reduction Strategies of the Iron and Steel Industry: An LCA-Based Study of Source, Process, End-Use and Cleaner Production Applications," Environmental Research, vol. 293, p. 123769, 2026, doi: 10.1016/j.envres.2026.123769.

[3] A. Al-Haddad and A. Abu-Rayash, "Volatility Modeling in Energy Commodity Markets: A Review of Responses to Shocks and the Path Towards Resiliency and Sustainability," Applied Energy, vol. 411, p. 127559, 2026, doi: 10.1016/j.apenergy.2026.127559.

[4] A. Sabzevari Zadeh, R. Sahraeian, and S. M. Homayouni, "A Dynamic Multi-Commodity Inventory and Facility Location Problem in Steel Supply Chain Network Design," International Journal of Advanced Manufacturing Technology, vol. 70, no. 5-8, pp. 1267-1282, 2014, doi: 10.1007/s00170-013-5358-2.

[5] M. A. Mohammadi, A. R. Sayadi, M. Khoshfarman, and A. Husseinzadeh Kashan, "A Systems Dynamics Simulation Model of a Steel Supply Chain—Case Study," Resources Policy, vol. 77, p. 102690, 2022, doi: 10.1016/j.resourpol.2022.102690.

[6] S. A. Hosseini Shekarabi, R. Kiani Mavi, and F. R. Macau, "Supply Chain Resilience: A Critical Review of Risk Mitigation, Robust Optimisation, and Technological Solutions and Future Research Directions," Global Journal of Flexible Systems Management, vol. 26, pp. 681-735, 2025, doi: 10.1007/s40171-025-00458-8.

[7] M. Parsaee, M. M. Mohtadi, and S. Khalili, "A model for assessing the resilience of supply chains using a combined DEMATEL, ANP, and grey relational analysis approach (Case study: Isfahan Steel Company)," Journal of Supply Chain Management, vol. 21, no. 65, pp. 60-72, 2020.

[8] S. H.-a. Mirghafoori, A. Morovati Sharifabadi, and F. Asadian Ardakani, "Evaluation of Suppliers Risk in Supply Chain Using Combining Fuzzy VIKOR and GRA Techniques," Industrial Management Journal, vol. 4, no. 2, pp. 153-178, 2012, doi: 10.22059/imj.2012.35444.

[9] S. H. R. Sadeghian and H. Mahdavi Molayemi, "Application of a Hybrid Model Based on Fuzzy Delphi, Fuzzy DEMATEL, and Fuzzy ANP for Supplier Selection in a LARG Supply Chain (Case Study: Khuzestan Steel Company)," Interdisciplinary Studies in Management and Engineering, vol. 5, no. 4, pp. 2302-2311, 2022. [Online]. Available: https://civilica.com/doc/1507413/.

[10] S. H. Abu Warda, N. H. A. F. Abu Ghararah, and M. A. F. Youssef, "The role of digital supply chain integration in improving product innovation: A field study on iron and steel companies in Egypt," Journal of Contemporary Commercial Studies, vol. 8, no. 14, pp. 1-61, 2022. [Online]. Available: https://csj.journals.ekb.eg/article_261229.html.

[11] N. Shahmohamadi, H. Moeinzad, S. Mehrinejad, and M. Keramati, "Presenting a Decision Support System Model based on the Analysis of Digital Supply Chain Management Risk Factors in the Country's Steel Industries," Dynamic Management and Business Analysis, vol. 2, no. 2, pp. 128-139, 2023, doi: 10.61838/dmbaj.2.2.10.

[12] M. Pourmehdi, M. M. Paydar, and E. Asadi-Gangraj, "Scenario-Based Design of a Steel Sustainable Closed-Loop Supply Chain Network Considering Production Technology," Journal of Cleaner Production, vol. 277, p. 123298, 2020, doi: 10.1016/j.jclepro.2020.123298.

[13] A. Khalili-Fard, F. Sabouhi, and A. Bozorgi-Amiri, "Data-Driven Robust Optimization for a Sustainable Steel Supply Chain Network Design: Toward the Circular Economy," Computers & Industrial Engineering, vol. 195, p. 110408, 2024, doi: 10.1016/j.cie.2024.110408.

[14] B. Figueiredo, R. B. Lopes, and A. Sousa, "Location-Routing Problems with Sustainability and Resilience Concerns: A Systematic Review," Logistics, vol. 9, no. 1, p. 12, 2025, doi: 10.3390/logistics9030081.

[15] Y. Wang, J. Wang, H. Zhang, and J. Song, "Bridging Prediction and Decision: Advances and Challenges in Data-Driven Optimization," Nexus, vol. 2, p. 100057, 2025, doi: 10.1016/j.ynexs.2025.100057.

[16] A. Anis Lahoud, A. S. Khan, E. Schaffernicht, M. Trincavelli, and J. A. Stork, "Predict-and-Optimize Techniques for Data-Driven Optimization Problems: A Review," Neural Processing Review, 2025, doi: 10.1007/s11063-025-11746-w.

[17] A. Wendt et al., "Identifying Bottlenecks in the Iron and Folic Acid Supply Chain in Bihar, India: A Mixed-Methods Study," BMC Health Services Research, vol. 18, no. 1, 2018, doi: 10.1186/s12913-018-3017-x.

[18] K. Ahmad et al., "Public Health Supply Chain for Iron and Folic Acid Supplementation in India: Status, Bottlenecks and an Agenda for Corrective Action Under Anemia Mukt Bharat Strategy," Plos One, vol. 18, no. 2, p. e0279827, 2023, doi: 10.1371/journal.pone.0279827.