SMART EOQ MODELS: INCORPORATING AI AND MACHINE LEARNING FOR INVENTORY OPTIMIZATION

Patel Nirmal Rajnikant; Ritu Khanna

doi:10.29121/ijoest.v9.i4.2025.709

Authors

Patel Nirmal Rajnikant Research Scholar, Faculty of Science, Department of Mathematics, Pacific Academy of Higher Education & Research University, Udaipur, Rajasthan, India
Dr. Ritu Khanna Professor & Faculty of Engineering, Pacific Academy of Higher Education & Research University, Udaipur, Rajasthan, India

DOI:

https://doi.org/10.29121/ijoest.v9.i4.2025.709

Keywords:

Dynamic Eoq, Reinforcement Learning, Stochastic Inventory Control, Perishable Inventory, Lstm Forecasting, Backorder Costs, Reorder Point Optimization, Supply Chain Resilience, Mathematical Inventory Models, Ai Operations

Abstract

Traditional Economic Order Quantity (EOQ) models rely on static assumptions (e.g., constant demand ????????, fixed holding cost ℎ), failing in volatile environments. This research advances dynamic inventory control through an AI-driven framework where:
1) Demand Forecasting: Machine learning (LSTM/GBRT) estimates time-varying demand:
????????ₜ = ???????? (????????ₜ; ????????) + ????????ₜ
(????????ₜ: covariates like promotions, seasonality; ????????ₜ: residuals)
2) Adaptive EOQ Optimization
Reinforcement Learning (RL) dynamically solves the following optimization problem: min????????????????,???????????????? ???????????????? ???????? (ℎ⋅????????????????++????????⋅????????????????−+????????⋅????????(????????????????))????
Subject to: ????????????????=????????????????−1+????????????????−????????????????
Where:
• Q_t: Order quantity at time t
• s_t: Reorder point at time t
• h: Holding cost per unit
• b: Backorder (shortage) cost per unit
• k: Fixed ordering cost
• δ(Q_t): Indicator function (1 if Q_t>0, else 0)
• I_t^+: Inventory on hand (positive part of I_t)
• I_t^-: Backordered inventory (negative part of I_t)
• D_t: Demand at time t
Validation was performed using sector-specific case studies.
• Pharma: Perishability constraint ????????ₜ⁺ ≤ ???????? (????????: shelf-life) reduced waste by 27.3%
• Retail: Promotion-driven demand volatility (????????²(????????ₜ) ↑ 58%) mitigated, cutting stockouts by 34.8%
Automotive: RL optimized multi-echelon coordination, reducing shortage costs by 31.5%.
The framework reduced total costs by 24.9% versus stochastic EOQ benchmarks. Key innovation: closed-loop control where ????????ₜ = RL(????????????????????????????????????????ₜ) adapts to real-time supply-chain states.

Downloads

Download data is not yet available.

References

Asad, M., & Zhang, L. (2022). Stochastic Inventory Optimization with Machine Learning Demand Estimators. Mathematics and Statistics Horizon, 16(2), 145–163. https://doi.org/10.1080/00207543.2022.2060308

Bakker, M., Riezebos, J., & Teunter, R. H. (2012). Review of Inventory Systems with Deterioration Since 2001. International Journal of Production Research, 50(24), 7117–7139. https://doi.org/10.1080/00207543.2011.613864

Bijvank, M., Vis, I. F. A., & Bozer, Y. A. (2014). Lost-Sales Inventory Systems with Order Crossover. European Journal of Operational Research, 237(1), 152–166. https://doi.org/10.1016/j.ejor.2014.01.050

Das, T., & Patil, S. (2020). Predictive Analytics in Inventory Systems: Neural Networks vs. ARIMA Models. Mathematics and Statistics Horizon, 14(4), 311–327. https://doi.org/10.1016/j.ijpe.2020.107724

F. W. Harris. (1913). How Many Parts to Make at Once. The Magazine of Management, 10(2), 135–136.

Ferreira, K. J., Lee, B. H. A., & Simchi-Levi, D. (2016). Analytics for an Online Retailer: Demand Forecasting and Price Optimization. Manufacturing & Service Operations Management, 18(1), 69–88. https://doi.org/10.1287/msom.2015.0561

Gijsbrechts, J., Boute, R. N., Van Mieghem, J. A., & Zhang, D. (2022). Can Deep Reinforcement Learning Improve Inventory Management? Management Science, 68(1), 243–265. https://doi.org/10.1287/mnsc.2020.3896

Govindan, K., Soleimani, H., & Kannan, D. (2020). Multi-Echelon Supply Chain Challenges: A Review and Framework. Transportation Research Part E: Logistics and Transportation Review, 142, 102049. https://doi.org/10.1016/j.tre.2020.102049

Gupta, N., & Kumar, D. (2023). EOQ Model Extensions with Reinforcement Learning For Real-Time Inventory Control. Mathematics and Statistics Horizon, 17(1), 67–83. https://doi.org/10.1109/TKDE.2022.3199987

Li, X., Shi, Y., & Wu, F. (2022). Ai-Driven EOQ Models: A Data-Driven Approach for Dynamic Demand Forecasting. Mathematics and Statistics Horizon, 16(4), 332–349. https://doi.org/10.1142/S0218001422500191

Oroojlooy, A., Nazari, M., Snyder, L. V., & Takác, M. (2020). A Deep Q-Network for the Beer Game: Reinforcement Learning for Inventory Optimization. INFORMS Journal on Computing, 32(1), 137–153. https://doi.org/10.1287/ijoc.2019.0891

Rossi, R. (2014). Stochastic Perishable Inventory Control: Optimal Policies and Heuristics. Computers & Operations Research, 50, 121–130. https://doi.org/10.1016/j.cor.2014.05.001

Salehi, F., Barari, A., & Fazli, S. (2021). Deep Learning Approaches for Supply Chain Inventory Optimization: An Empirical Evaluation. Mathematics and Statistics Horizon, 15(3), 212–230. https://doi.org/10.1016/j.asoc.2020.106684

Scarf, H. (1960). The Optimality of (s, S) Policies in the Dynamic Inventory Problem. In K. Arrow, S. Karlin, & P. Suppes (Eds.), Mathematical Methods in the Social Sciences (pp. 196–202). Stanford University Press.

Schmitt, A. J., Kumar, S., & Gambhir, S. (2017). The Value of Real-Time Data in Supply Chain Decisions: Limits of Static Models in a Volatile World. International Journal of Production Economics, 193, 684–697. https://doi.org/10.1016/j.ijpe.2017.08.017

Seaman, B. (2021). Time Series Forecasting with LSTM Neural Networks for Retail Demand. Expert Systems, 38(4), e12609. https://doi.org/10.1111/exsy.12609

Singh, R., Mishra, A., & Ramachandran, K. (2021). Hybrid EOQ Forecasting Models Combining LSTM and Gradient Boosted Trees. Mathematics and Statistics Horizon, 15(1), 39–58. https://doi.org/10.1016/j.cor.2020.105251

Swaminathan, J. M. (2023). Intelligent Decision Support for Inventory Optimization using Machine Learning. Mathematics and Statistics Horizon, 17(2), 101–117. https://doi.org/10.1007/s00199-023-01473-6

Trapero, J. R., Kourentzes, N., Fildes, R., & Spiteri, M. (2019). Promotion-Driven Demand Forecasting in Retailing: A Machine Learning Approach. International Journal of Forecasting, 35(2), 712–726. https://doi.org/10.1016/j.ijforecast.2018.11.005

Wang, J., & Li, X. (2023). A Comparative Study of AI Algorithms in EOQ Models with Backordering. Mathematics and Statistics Horizon, 17(3), 198–213. https://doi.org/10.1016/j.eswa.2023.120012

Zipkin, P. H. (2000). Foundations of Inventory Management. McGraw-Hill.