Advancing Sustainable Supply Chains Through Knowledge Graph Completion and Graph-Based Artificial Intelligence

limited supply-chain visibility: Modern supply chains lack sufficient visibility into upstream relationships, environmental risks, and ethical sourcing practices needed for sustainability targets.

incomplete sparse supply-chain data: Sustainability-oriented supply chain reasoning must operate over incomplete and sparse data in complex supply chain settings.

limited supply-chain visibility: Modern supply chains lack sufficient visibility into upstream relationships, environmental risks, and ethical sourcing practices needed for sustainability targets.

incomplete sparse supply-chain data: Sustainability-oriented supply chain reasoning must operate over incomplete and sparse data in complex supply chain settings.

knowledge graph completion for supply chains: The paper proposes an AI-based approach using knowledge graph completion and link prediction to support sustainability-oriented supply chain decision-making.

multi-relational supply chain knowledge graph: The approach constructs a multi-relational supply chain knowledge graph representing suppliers, products, certifications, locations, and their relationships.

graph neural link and attribute inference: The paper applies graph neural networks to infer missing supply chain links and sustainability-related attributes.

heterogeneous graph reasoning framework: The framework combines relational structure with inductive learning to generate interpretable recommendations under uncertainty.

knowledge graph completion for supply chains: The paper proposes an AI-based approach using knowledge graph completion and link prediction to support sustainability-oriented supply chain decision-making.

supports feasibility-oriented decisions:

real-world dataset applicability:

practical sustainability-aware decision support foundation: