GraphRAG (Graph-Based Retrival Augmented Generation) Application in UAM System of Systems

Urban Air Mobility (UAM) aims to transform congested urban transportation by deploying electric vertical takeoff and landing (eVTOL) vehicles with integrated air traffic management and infrastructure. This network of diverse, interacting components classifies UAM as a System of Systems (SoS). The integration of independently operated subsystems within UAM present significant challenges, including operational independence, data interoperability, dynamic evolution, and emergent behaviors. This paper introduces how GraphRAG, an open-sourced hybrid approach combining knowledge graphs with Retrieval-Augmented Generation (RAG), could address these challenges by enhancing data integration and facilitating real-time updates from unstructured sources. Unlike traditional methods, GraphRAG leverages structured, semantic representations to enable efficient multi-level querying, reduce redundancy, and dynamically adapt to new information. Practical application scenarios, including interdisciplinary knowledge integration, dynamic regulatory compliance and the combination with MBSE, further showcase the system's capabilities to assist decision making in complex SoS environment. Future research is suggested to explore deeper integration with ontology frameworks, enhancing semantic consistency and automated discrepancy detection across diverse data sources.