Methodology to Identify Physical or Computational Experiment Conditions for Uncertainty Mitigation

The design and analysis of complex engineering systems rely heavily on the integration of simulations across multiple subsystems to drive design decisions. This paper introduces a methodology for designing computational or physical experiments aimed at mitigating system-level uncertainties, particularly epistemic uncertainties, which in turn reduces risk, potential negative consequences impacting performance, cost, and safety. By leveraging a structured ”problem ontology,” global sensitivity analysis, and tailored experiment design, the methodology provides a systematic framework to identify and reduce epistemic uncertainties in engineering design. The problem ontology organizes the relationships among physical, functional, and modeling architectures, enabling targeted identification of critical uncertainties through sensitivity analysis. Computational and physical experimentation are central to the framework, enabling targeted knowledge generation for uncertainty mitigation. This framework combines a top-down system and requirements decomposition (ontology) with sensitivity-driven experiment design. It is demonstrated through a case study on an early-stage-design blended-wing-body aircraft concept, showcasing how aerostructures analysis are used to mitigate system-level uncertainty, by computer experiments or guiding physical experimentation. This versatile framework is applicable across various design challenges, enabling risk-informed decision-making.