What question did this study set out to answer?

The main aim is to identify high-risk nodes and abnormal behaviors in dynamic supply chain networks using advanced modeling techniques.

January 18, 2026Engineering ReportsOpen Access

TG ‐ RRNet : A Supply Chain Risk Perception Network Integrating Temporal Modeling and Generative Anomaly Detection

View Full Paper

Ask AI

Bookmark

Authors

This framework demonstrates improved high-risk node identification and anomaly detection in supply chains, highlighting practical applications.

The main aim is to identify high-risk nodes and abnormal behaviors in dynamic supply chain networks using advanced modeling techniques.
Developed a heterogeneous dynamic graph to capture structural evolution over time.
Utilized a temporal-aware graph neural network to extract dynamic node representations.
Employed a generative anomaly detection module via variational autoencoders.
Applied a multimodal cross-attention mechanism for unified risk representation.
TG-RRNet significantly improved high-risk node identification compared to existing methods.
Demonstrated enhanced anomaly detection capabilities, effectively identifying 'camouflaged normal' nodes.
Achieved strong generalization across real-world supply chain datasets.

Ping Li (2026) studied this question.

View Full Paper

Ask AI

Bookmark

This framework demonstrates improved high-risk node identification and anomaly detection in supply chains, highlighting practical applications.