Work

Monte Carlo uncertainty quantification and sensitivity analysis for the C5G7 benchmark

Year: 2025

Type: report

Abstract: Overview • Widely used to quantify variance in model output and attribute it to different uncertainty sources • Global: sample all parameters across their entire (respective) distributions • Sidenote -Not the same as local sensitivity analysis!Variance-based global sensitivity analysis (GSA) • Output has some associated uncertainty due to solver noise • Solver noise complicates existing GSA methods Stochastic computational models Goal: variance-based GSA with stochastic models without having to resolve the solver uncertainty to (near) zero 2

DOI: 10.2172/3030918

Authors: Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States), Kayla Clements, National Nuclear Security Administration, Todd Palmer, Aaron Olson, Gianluca Geraci, Ilham Variansyah

Institutions: Oregon State University, Sandia National Laboratories

Language: English

Topic: Probabilistic and Robust Engineering Design

Subfield: Statistics, Probability and Uncertainty

Field: Decision Sciences

Domain: Social Sciences

Open Access status: gold

OpenAlex ID: W7155617000

Cites:

Cited by:

Related to: 10