Extending Architecture Frameworks for Automated Operations Research Analysis in System of Systems

The increasing complexity and interdependence of Systems of Systems (SoS) demand innovative approaches for optimizing their performance. This paper addresses the challenge of enhancing the quantitative analytical capabilities of architecture frameworks (AFs) by integrating them with operations research (OR) methodologies. Specifically, we propose a metamodel-based approach that automates OR analysis within AFs, focusing on the Unified Architecture Framework (UAF). The method employs extended metamodels and a validated OR library, enabling automated optimization and analysis. Additionally, the integration of OR Large Language Models (LLMs) allows the handling of optimization scenarios that fall outside the scope of traditional OR libraries. The approach is demonstrated through a case study on a drone swarm delivery SoS, where it enhances modeling efficiency and facilitates automated problem-solving. This work contributes to the advancement of AFs for early-stage decision support in complex SoS environments, offering significant improvements in flexibility, automation, and applicability.