Abstract

Carrying costs which include inventory holding, obsolescence, and lost opportunity costs are a big obstacle to improving efficiency in multi-tier supply chain networks. This paper proposes an AI-enhanced Vendor Managed Inventory (VMI) system, backed by a causal inference framework, to solve this problem by optimizing inventory levels and reducing costs. Machine learning is employed for demand forecasting, whereas decision-making for replenishment is done by reinforcement learning. Furthermore, the framework uses causal methods like propensity score matching (PSM) and difference-in-differences (DiD) to measure the effect of AI intervention. A simulation experiment of a three-tier supply chain (supplier manufacturer retailer) demonstrates how a 20-30% reduction in carrying costs might be accomplished. The approach solves the issue of supply chain data endogeneity and provides manufacturing as well as retail sectors with reusable knowledge. The findings of this paper show that AI optimization along with causal analysis constitute a potent method for launching green supply chain management.

References

• Disney, S. M., & Towill, D. R. (2003). Vendor-managed inventory and bullwhip reduction in a two-level supply chain. International Journal of Operations & Production Management.
• Wallenburg, C. M., et al. (2020). Vendors managed inventory in multi-tier supply chains. Journal of Business Logistics.
• Chopra, S., & Meindl, P. (2016). Supply chain management: Strategy, planning, and operation. Pearson.
• Toorajipour, R., et al. (2021). Artificial intelligence in supply chain management: A systematic literature review. Journal of Business Research.
• Brintrup, A., et al. (2023). Supply chain visibility and resilience with AI. International Journal of Production Research.
• Rolf, B., et al. (2023). A review on reinforcement learning algorithms and applications in supply chain management. International Journal of Production Research.
• Lee, H. L., et al. (1997). Information distortion in a supply chain: The bullwhip effect. Management Science.
• Simchi-Levi, D., et al. (2021). Designing and managing the supply chain. McGraw-Hill.
• Nahmias, S., & Olsen, T. L. (2015). Production and operations analysis. Waveland Press.
• Boute, R. N., et al. (2022). Deep reinforcement learning for inventory management. Manufacturing & Service Operations Management.
• Gijsbrechts, J., et al. (2023). An application of deep reinforcement learning and vendor-managed inventory in perishable supply chain management. Engineering Applications of Artificial Intelligence.
• Ivanov, D. (2021). Digital supply chain management with AI. Computers & Industrial Engineering.
• Ni, D., et al. (2020). Machine learning in supply chain management: A review. Transportation Research Part E.
• Boute, R. N., & Van Mieghem, J. A. (2021). Global dual sourcing: Tailored base-surge policies. Operations Research.
• Gijsbrechts, J., et al. (2023). Deep reinforcement learning in VMI. ScienceDirect.
• Kim, J., et al. (2021). A deep reinforcement learning approach for optimal replenishment policy in a vendor managed inventory setting for semiconductors. IEEE Winter Simulation Conference.
• Bertsimas, D., et al. (2024). Generative AI for supply chain scenario planning. arXiv preprint.
• McKinsey & Company. (2023). Succeeding in the AI supply chain revolution.
• Pearl, J., & Mackenzie, D. (2018). The book of why: The new science of cause and effect. Basic Books.
• Cui, R., et al. (2022). Causal inference in operations management. Manufacturing & Service Operations Management.
• Bodendorf, F., et al. (2022). A mixed methods approach to analyze and predict supply disruptions by combining causal inference and deep learning. Computers & Industrial Engineering.
• Wyrembek, M., et al. (2024). What if? Causal machine learning in supply chain risk management. arXiv.
• Gottimukkala, S. R., et al. (2025). Prescriptive analytics for next-gen supply chains: Integrating causal AI with digital twin technologies. IEEE.
• Ameer, T., & Fatahi Valilai, O. (2025). Cloud-native causal AI for supply chain KPI monitoring: A GCP framework to diagnose out-of-stock events. Machine Learning with Applications, 100765. https://doi.org/10.1016/j.mlwa.2025.100765
• Kaleva, H., & Småros, J. (2023). The complete guide to machine learning in retail demand forecasting. RELEX Solutions (technical report).
• Feizabadi, J. (2022). Machine learning demand forecasting and supply chain performance. International Journal of Logistics Research and Applications, 25(2), 119–142.
• Lu, Y., Jomhari, N., Liao, S., & Goyal, S. B. (2025). Warehouse demand variation prediction and optimization based on machine learning algorithms. In Proceedings of the 4th Asia-Pacific Artificial Intelligence and Big Data Forum (AIBDF 2024). ACM. https://doi.org/10.1145/3718491.3718674
• Nair, D., & Saenz, M. J. (2024). Pair people and AI for better product demand forecasting. MIT Sloan Management Review.
• Disney, S. M., et al. (2023). Multi-echelon inventory optimization using deep reinforcement learning. Central European Journal of Operations Research.
• Boute, R. N., et al. (2022). Can deep reinforcement learning improve inventory management? Manufacturing & Service Operations Management.
• Ivanov, D. (2024). AI-driven supply chain resilience. International Journal of Production Economics.
• Wyrembek, M., et al. (2024). Causal machine learning for supply chain interventions. arXiv.
• Various. (2025). Limitations of simulation in SCM research. Journal of Operations Management.
• European Commission. (2025). Policy incentives for AI in sustainable supply chains.
• Pournader, M., et al. (2021). Artificial intelligence in supply chains: Review and future directions. International Journal of Production Economics. Culot, G., et al. (2020). Behind the definition of Industry 4.0: Analysis and open questions. International Journal of Production Economics.