Predictive analytics for supply chain resilience in urban infrastructure networks using graph convolutional networks
Yanfang Liu, Ruijun Hu, Wen Zhang, Qian Wu · 2026
Read Predictive analytics for supply chain resilience in urban infrastructure networks using graph convolutional networks by Yanfang Liu, Ruijun Hu, Wen Zhang, Qian Wu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0345444.txt.
Description
Yanfang Liu and three colleagues build a system for predicting infrastructure disruption in New York City in this 2026 study. Urban supply chains depend on networks of roads and facilities whose failure cascades quickly. Their framework combines network analysis, a graph-based neural network that learns from the city's structure itself, and stress testing. It was trained on thousands of infrastructure nodes, nearly one hundred thousand recorded disruption events, and a large traffic flow archive.
The graph model predicted flow disruption far more accurately than four established forecasting baselines, and removing the network structure inflated its errors by nearly a third, proving the map itself carries information. Stress tests showed targeted attacks on hub nodes fragment the network fast, though random failures overtake them in total damage beyond a point. Past flow patterns and timing carried most of the predictive weight. The authors flag dual-risk nodes for priority protection.
Catalog record
| CREATOR | Yanfang Liu, Ruijun Hu, Wen Zhang, Qian Wu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 89118 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |