Uncertainty Quantification Methods for Model Calibration, Validation, and Risk Analysis


In this paper we propose a series of methodologies to address the problems in the NASA Langley Multidisciplinary UQ Challenge. A Bayesian approach is employed to characterize and calibrate the epistemic parameters in problem A, while variance-based global sensitivity analysis is proposed for problem B. For problems C and D we propose nested sampling methods for mixed aleatory-epistemic UQ.

AIAA-2014-1497, 16-th AIAA Non-Deterministic Approaches Conference