On Methods of Uncertainty Quantification in Computational Modeling and Simulation: Harmonizing Modeling and Metrology Practices

Author(s)

Abstract

As Computational Modeling and Simulation (CMS) increasingly drives critical engineering and regulatory decisions, establishing model credibility has become paramount. This report presents a comprehensive approach to harmonizing the uncertainty quantification (UQ) methods within the computational modeling standards of the American Society of Mechanical Engineers (ASME) with the empirical metrology practices of the Joint Committee for Guides in Metrology (JCGM). We explain a standardized, risk-based approach to credibility assessment, demonstrating how Verification, Validation, and Uncertainty Quantification (VVUQ) efforts can be scaled in proportion to model risk, which is determined by model influence and decision consequence. To navigate the modern CMS landscape, we categorize models across creation, logic, and transparency perspectives, contrasting mechanistic first-principles approaches with empirical data-driven techniques. Furthermore, this work details the mathematical methods of UQ, comparing the propagation of variances against the propagation of distributions, and highlights recent developments in the physical interpretation of adjoint methods to overcome computational bottlenecks in sensitivity analysis. Finally, we look ahead to the challenges of UQ in Artificial Intelligence (AI) models and the integration of UQ practices into digital twin ecosystems and industrial ontologies. Ultimately, this work provides a unified framework for bridging the gap between computational predictions and physical reality.

Download Paper

Keywords

Manufacturing

Citation

Issues

If you have any questions about this publication or are having problems accessing it, please contact [email protected].