Predicting Supply Shocks Before They Propagate Using Temporal Graph Reasoning
International Journal of Advanced Engineering and Technology Research · 12.05.2026 · 2 (1) · 6-12
DOI: 10.54097/k79vxk50 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Engineering and Technology Research, online erschienen am 12.05.2026. Verfasst von Zeyu Qin und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Supply shocks in modern production networks rarely originate from isolated events; they propagate upstream and downstream through densely interconnected supplier-customer relationships, often manifesting far from their point of origin before conventional monitoring systems detect any signal. This paper proposes a temporal graph reasoning framework for predicting supply shocks before they complete their propagation cascade. By modeling supplier-customer networks as dynamic directed graphs with time-stamped transaction edges, the proposed architecture integrates diffusion-based spatial aggregation with recurrent temporal encoding to capture the evolving topology of inter-firm dependencies. A graph convolutional network (GCN) backbone encodes node-level structural features, while a long short-term memory (LSTM) layer learns the sequential dynamics of nodal activity. The model is evaluated against static graph baselines and non-graph sequential models on a multi-industry transaction dataset spanning 36 months. Results demonstrate that the proposed approach achieves a mean absolute error (MAE) reduction of 18.3% over the best competing baseline and exhibits substantially earlier detection capability, identifying disruption precursors an average of 11.3 days before shock propagation reaches tier-1 suppliers. These findings suggest that temporal graph reasoning constitutes a promising direction for proactive supply chain risk management under conditions of structural uncertainty.
Zitieren
APA
Qin, Z., Gu, Y., & Foster, D. (2026). Predicting Supply Shocks Before They Propagate Using Temporal Graph Reasoning. International Journal of Advanced Engineering and Technology Research, 2(1), 6-12. https://doi.org/10.54097/k79vxk50
Vancouver
Qin Z, Gu Y, Foster D. Predicting Supply Shocks Before They Propagate Using Temporal Graph Reasoning. International Journal of Advanced Engineering and Technology Research. 2026;2(1):6-12. doi:10.54097/k79vxk50
BibTeX
@article{Qin2026,
author = {Qin, Zeyu and Gu, Yuxin and Foster, Daniel},
title = {Predicting Supply Shocks Before They Propagate Using Temporal Graph Reasoning},
journal = {International Journal of Advanced Engineering and Technology Research},
year = {2026},
volume = {2},
number = {1},
pages = {6--12},
doi = {10.54097/k79vxk50}
}
EUVIMED