A Graph Neural Network (GNN) based system for supply chain risk assessment and early warning. This project leverages multi-layer graph architectures and ensemble GNN models (GCN, GAT, and Temporal GNN) to predict supplier risks, simulate disruption cascades, and provide actionable insights for supply chain resilience.