Supply Chain Risk

Low-code analytic pipelines that surface concealed connections in supply-chain data.

Analysts assessing supply-chain risk work across data that is large, heterogeneous, and sometimes deliberately obscured, the "adversarial capital" problem, where a foreign investor's connection to an acquisition target is concealed behind aliases and intermediaries. Building the analytic pipeline to surface those connections has been out of reach for most analysts. With ARLIS at the University of Maryland, Fraunhofer USA stood up a rapid-experimentation infrastructure to test whether modular, low-code pipelines could close that gap, and instantiated it against four CFIUS-style case studies. The resulting prototype dashboard, built on a Neo4j knowledge graph with NLP-based ingestion and serverless AWS components, implements three analyst-facing concepts, aggregate and explore, tipping and cueing, and on-demand enrichment, letting an analyst compose complex graph queries from prebuilt modules rather than writing them.