New Energy Vehicle Closed-Loop Supply Chain Optimization Considering Carbon Footprint and Supply Chain Disruption Risks: An Innovative Method Integrating Robust Optimization and Machine Learning

DOI: https://doi.org/10.62517/jike.202604131

Author(s)

Yu Wang, Weilun Huang

Affiliation(s)

School of Finance and Trade, Wenzhou Business College, Wenzhou, China

Abstract

Faced with the dual impacts of tightening global carbon neutrality policies and frequent supply chain disruptions, the closed-loop supply chain (CLSC) of new energy vehicles (NEVs) has become the core carrier for balancing industrial growth and sustainable development. However, existing operations research (OR) studies in this field have significant limitations: isolated multi-objective optimization, shallow integration between machine learning and optimization models, etc. This paper proposes an innovative methodology integrating Long Short-Term Memory (LSTM) network and robust optimization, which realizes the coordinated optimization of three objectives—carbon footprint, disruption risk, and economic cost—through a closed-loop mechanism of "dynamic prediction-adaptive optimization-real-time feedback". Taking the supply chain of Volkswagen Group's European MEB platform as an empirical case, and based on public data and industry reports for verification, the results show that the proposed Model-0L integrating LSTM and robust optimization is significantly superior to traditional models in key indicators such as total operational cost and stockout rate.

Keywords

New Energy Vehicles (NEVs); Closed-Loop Supply Chain (CLSC); Carbon Footprint; Supply Chain Disruption; Robust Optimization; Long Short-Term Memory (LSTM); Operations Research (OR)

References

[1]European Commission. Regulation (EU) 2023/1542 on Batteries and Waste Batteries. Brussels: European Commission; 2024. [2]European Commission. (2024). Carbon Border Adjustment Mechanism (CBAM) Implementation Guidelines 2026. Brussels: European Commission. [3]Volkswagen Group. (2024). Sustainability Report 2024. Wolfsburg: Volkswagen AG. [4]London Metal Exchange (LME). (2024). Battery Metals Price & Volatility Report 2024. London: LME. [5]United States Geological Survey (USGS). (2024). Mineral Commodity Summaries 2024. Reston: USGS. [6]International Energy Agency (IEA). (2024). Global EV Outlook 2024. Paris: IEA. [7]Shen, L., Liu, Z., & Li, Y. (2023). Carbon-constrained closed-loop supply chain network design with uncertain demands and supply disruptions. European Journal of Operational Research, 309(2), 678-692. [8]Tu, S. (2025). Enlightenments from the Rise and Fall of the Red Sea: On the First Anniversary of the Red Sea Crisis. China Ports, (02), 20-22. [9]Guide, V. D. R., Jayaraman, V., & Srivastava, R. (2003). Supply chain models for recycled materials. Management Science, 49(1), 102-117. [10]Wang, H., Chen, J., & Zhang, H. (2022). Machine learning-aided demand forecasting for sustainable supply chains. European Journal of Operational Research, 298(3), 954-968. [11]Liu, J. (2022). Order Decision and Contract Design Based on Distributionally Robust Optimization with Random Yield, Demand and Price (Doctoral Dissertation). South China University of Technology, Guangzhou. [12]Salinas, D., Flunkert, V., & Gasthaus, J. (2019). DeepAR: Probabilistic forecasting with autoregressive recurrent networks. Advances in Neural Information Processing Systems, 32, 11387-11397. [13]Zhou, Y., Li, Q., & Zhao, Y. (2025). Research on the Full-Life-Cycle Standard System of Power Batteries. China Standardization, (15), 102-105. [14]Fang, S., Liu, Z., & Cheng, Y. (2025). Integration of Machine Learning Methods and Enterprise Management Research: Review and Prospect. Economic Management, 47(03), 183-208. [15]Ling, Q. Y. (2025). Research on the Impact of the EU's New Battery Regulation on China's New Energy Vehicle Exports. In Proceedings of the Conference on New Quality Productive Forces and the Development of Emerging Economies (Volume 1) (pp. 374-383). Foshan University Press. [16]European Energy Exchange (EEX). (2024). EU ETS Price Statistics 2024. Leipzig: EEX. [17]Du, H. Y., & Dai, H. S. (2025). Patient Capital, Industrial Chain and Supply Chain Resilience, and High-Quality Development of Manufacturing Enterprises. Mathematics in Practice and Theory, 1-19. [18]Yang, Y., Zang, Z., & Cheng, L. (2025). Robust Optimal Operation Strategy of Active Distribution Network Based on "Prediction-Control-Correction" Closed-Loop Feedback. Proceedings of the CSEE, 1-14. [19]Xia, M. (2019). Research on Inventory Control Optimization of Closed-Loop Supply Chain with Uncertain Recovery Rate (Master's Dissertation). Nanjing University of Posts and Telecommunications, Nanjing. [20]Fang, H. (2025). Seeking the "Pareto Optimal Solution" for the High-Quality Development of New Energy. People's Political Consultative Conference News, (005). [21]Wu, J., & Wang, S. F. (2024). In-Depth Analysis of the EU's New Battery Regulation. Shanghai Auto, (07), 45-48+53.