Addressing the issues of retail stockouts leading to truncated sales and the inability of fixed-probability replenishment to promptly reflect SKU inventory risk, this study examines the incremental decision-making value of inventory status information based on probabilistic demand forecasting. Using FreshRetailNet-50K data, the study employs TimesNet to recover potential demand and utilizes LightGBM quantile regression to predict demand during the protection period, classifying states into shortage, healthy, and excess. Under the same forecasting environment, it compares health-triggered and probability-driven replenishment. Results show that the three states maintain a stable risk ranking; compared to B2, the proposed fill rate increases by 0.37 percentage points, average inventory increases by 1.60%, and the inventory difference after service matching is −0.06 NDU, with confidence intervals crossing zero. The incremental effect mainly occurs in SKUs with high volatility and long lead times. This study demonstrates that inventory health information has conditional replenishment value and can serve as an interpretable risk calibration interface connecting probabilistic forecasting and replenishment rules.

Qian Lin. Research on SKU-Level Inventory Health Monitoring and Dynamic Replenishment Strategies. International Journal of Big Data Intelligent Technology (2026), Vol. 7, Issue 2: 105-115. https://doi.org/10.38007/IJBDIT.2026.070211.

[1] Trapero J R, Holgado de Frutos E, Pedregal D J. Demand forecasting under lost sales stock policies[J]. International Journal of Forecasting, 2024, 40(3): 1055-1068. DOI: 10.1016/j.ijforecast.2023.09.004.

[2] Theodorou E, Spiliotis E, Assimakopoulos V. Forecast accuracy and inventory performance: Insights on their relationship from the M5 competition data[J]. European Journal of Operational Research, 2025, 322(2): 414-426. DOI: 10.1016/j.ejor.2024.12.033.

[3] Hasni M, Babai M Z, Rostami-Tabar B. A hybrid LSTM method for forecasting demands of medical items in humanitarian operations[J]. International Journal of Production Research, 2024, 62(17): 6046-6063. DOI: 10.1080/00207543.2024.2306904.

[4] Wang S, Kang Y, Petropoulos F. Combining probabilistic forecasts of intermittent demand[J]. European Journal of Operational Research, 2024, 315(3): 1038-1048. DOI: 10.1016/j.ejor.2024.01.032.

[5] Van der Haar J F, Wellens A P, Boute R N, Basten R J I. Supervised learning for integrated forecasting and inventory control[J]. European Journal of Operational Research, 2024, 319(2): 573-586. DOI: 10.1016/j.ejor.2024.07.004.

[6] Wang, B. (2025). Methods of Load Optimization for Computer Systems Based on Physical Principles.

[7] Guo, X. (2025, March). Research on Blockchain-Based Financial AI Algorithm Integration Methods and Systems. In 2025 IEEE International Conference on Electronics, Energy Systems and Power Engineering (EESPE) pp.816-821. IEEE.

[8] Guo, X. (2025, April). Research on Financial Trading Algorithms Based on Deep Reinforcement Learning. In 2025 IEEE 3rd International Conference on Control, Electronics and Computer Technology (ICCECT) pp.1711-1715. IEEE.

[9] Yu, X. (2026). Exploration of Multi-Channel Conversion Path Optimization Methods Based on A/B Testing.

[10] Jing, X. (2025, November). In-Depth Analysis of the Optimization Decision Model of Financial Investment Portfolio Based on Machine Learning Algorithms in Python Scikit-Learn Library. In International Conference on Computational Technologies for Research in Data Analytics pp.130-140. Cham: Springer Nature Switzerland.

[11] Liu, X. (2026). Research on the Application of Artificial Intelligence in Consumer Privacy Data Protection. Procedia Computer Science, 279, 956-965.

[12] Liu, X. (2026). Construction of Consumer Data Privacy Protection System Based On Blockchain Technology. Procedia Computer Science, 282, 2082-2091.

[13] Huijie Pan. Discussion on Low-Latency Computing Strategies in Real-Time Hardware Generation. International Journal of Neural Network (2025), Vol. 4, Issue 1: 57-64.

[14] Han, W. (2025, November). An Empirical Study on Building a High-precision Forecasting System for Financial Market Trends Based on the PyTorch Deep Learning Framework. In International Conference on Computational Technologies for Research in Data Analytics (pp. 518-528). Cham: Springer Nature Switzerland.

[15] Yuan, Y. (2026). Research on Memory Management and Dynamic Reasoning Methods for Intelligent Agents Based on Large Language Models.

[16] Yang, Y. (2026). Causal Inference-based Identification of Incremental Effects in Digital Advertising and Optimization Pathways for Resource Allocation. Advances in Computer and Communication, 7(2).

[17] Sun, L. (2026). Consistency Optimization Strategies for Distributed Advertising Systems Under Cross-regional Deployment. Advances in Computer and Communication, 7(2).PK0977