Dynamic Heterogeneous Graph Learning and Task-Conditioned Meta-Learning for Cold-Start Demand Forecasting in Multi-Warehouse Supply Chains
Mathematics · 17.08.2026 · 14 (16) · 2971
DOI: 10.3390/math14162971 ↗Kurz gefasst
Fachartikel in der Zeitschrift Mathematics, online erschienen am 17.08.2026. Verfasst von Lei Ni und 2 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Scientific Research Startup Foundation of Guang’an Institute of Technology.
Abstract (Originalsprache)
New products have little sales history, which makes short-horizon demand forecasting difficult. This study tests whether a product–store–time graph can provide pre-origin relational context and whether task-conditioned meta-learning can adapt a forecasting model from limited support observations. IGEML combines a dynamic heterogeneous graph encoder with a task-conditioned model-agnostic meta-learning adapter. The cold-start graph contains product, store, and time-window nodes; user–product interactions are excluded at the forecast origin. In the 14-day evaluation, IGEML achieves MSE = 0.2345, MAE = 0.1532, and MAPE = 0.1206. In the k = 1 setting, IGEML achieves MAE = 0.1269, with an online adaptation time of 2.8 ms per SKU. The framework defines product-level forecasting tasks, maps engineered features to graph and adaptation modules, and controls the information available at the forecast origin. Conditioning task adaptation on pre-origin relational information enables cold-start forecasting without target-period demand or post-origin user interactions.
Förderung
- Scientific Research Startup Foundation of Guang’an Institute of Technology
Zitieren
APA
Ni, L., Fu, N., & Huang, Z. (2026). Dynamic Heterogeneous Graph Learning and Task-Conditioned Meta-Learning for Cold-Start Demand Forecasting in Multi-Warehouse Supply Chains. Mathematics, 14(16), 2971. https://doi.org/10.3390/math14162971
Vancouver
Ni L, Fu N, Huang Z. Dynamic Heterogeneous Graph Learning and Task-Conditioned Meta-Learning for Cold-Start Demand Forecasting in Multi-Warehouse Supply Chains. Mathematics. 2026;14(16):2971. doi:10.3390/math14162971
BibTeX
@article{Ni2026,
author = {Ni, Lei and Fu, Ning and Huang, Zhonglin},
title = {Dynamic Heterogeneous Graph Learning and Task-Conditioned Meta-Learning for Cold-Start Demand Forecasting in Multi-Warehouse Supply Chains},
journal = {Mathematics},
year = {2026},
volume = {14},
number = {16},
pages = {2971},
doi = {10.3390/math14162971}
}
EUVIMED