PBESProceedings of Business and Economic Studies2209-26412209-265XBio-Byword Scientific Publishing Pty. Ltd.10.26689/pbes.v9i6.15468ArticleData Island and Algorithm Fragmentation Dilemmas in Supply Chain Demand Forecasting and Federated Learning Optimization Pathwayshttps://artdesignp.com/journal/PBES/9/6/10.26689/pbes.v9i6.15468XueJunhua2026962026-07-15In supply chain demand forecasting, data is scattered and stored in various enterprise nodes, forming data islands. The independent operation of prediction algorithms by each node is not coordinated, resulting in algorithm fragmentation. The interaction between the two reinforces the amplification of prediction bias, the intensification of the bullwhip effect, and resource mismatch. Traditional centralized modeling is difficult to implement due to data privacy and commercial confidentiality constraints. Federated learning, as a distributed machine learning paradigm, allows models to train collaboratively without moving data, breaking data silos through cross-node parameter aggregation, and unifying algorithm logic and alleviating algorithm fragmentation through iterative mechanisms of global and local models. Embedded differential privacy and security aggregation technologies, while protecting sensitive enterprise information, achieve joint modeling, simulation validation shows that this path reduces prediction error by 22% to 30% within a controllable accuracy loss range, thereby providing a decentralized solution for supply chain demand forecasting that balances both security and collaboration.

Noor M, Fei W, Ullah K, et al., 2026, Harnessing Machine Learning for Supply Chain Management: A Systematic Review and Future Research Framework. Computers and Electrical Engineering, 135: 111189.

Prastuti M, Pujawan NI, Widodo E, et al., 2026, A Predictive Analytics Method for Multiregional Supply Chain Demand Forecasting with Spatial Time Series and Machine Learning. Decision Analytics Journal, 19: 100706.

Ji Y, Su K, Chen L, 2026, Research on a Lightweight Supply Chain Demand Forecasting Model Based on STL Decomposition and Dynamic Feature Selection. Forum on Research and Innovation Management, 4(4): 33.

Xu K, Liu J, 2026, A Puzzle-Based BiLSTM Model for Accurate Supply Chain Demand Forecasting: Regular Section. IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, E109.A(4): 818–828.

Ding J, Li Z, Lee SE, 2026, Seeing the Unseen: AI Assimilation and Supply–Demand Visibility for Effective Risk Management in Manufacturing Supply Chains. Systems, 14(3): 300.

Bao S, 2026, Research on Supply Chain Inventory Demand Forecasting Model Based on LightGBM. GBP Proceedings Series, TAPEE2025, 86–92.

Shen F, Yang Z, Zhao Z, et al., 2025, Integration of Machine Learning-Based Demand Forecasting and Economic Optimization for the Natural Gas Supply Chain. Energies, 18(23): 6172.

Masum MKA, Azad KAM, Bhuiyan ATM, et al., 2026, Bridging Data Silos in Corporate Governance: A Hierarchical Stacking Ensemble With Federated Dynamic Aggregation. Applied Computational Intelligence and Soft Computing, 2026(1): 8836162.

Gao B, Wu P, 2026, Design and Implementation of Non-Intrusive Logistics System Architecture for Data Silo Integration. Engineering Research Express, 8(3): 035230.

Álvarez RDJ, 2026, Beyond Data Silos: Federated Governance as Legal Response to Planetary Ecological Collapse. International Environmental Agreements: Politics, Law and Economics, (prepublish): 1–21.