NAFEMS-SWG Benchmark Problem 2022 - Uncertain Knowledge

This is an open repository containing the codes used to tackle the challenge problems. Details to the challenge problems can be found on the PDF document titled: "benchmark_april_2022_article.pdf".

Challenge Problem 1:

Run the MATLAB code named "NAFEMS_Problem_1.m" followed by the R code named "NAFEMS1.R"

Challenge Problem 2:

For the log model evidence analysis, run the MATLAB files "Cluster1.m" and "Cluster2.m" to run 100 simulations of the Bayesian model updating for the Scaled Beta and Normal distribution models respectively. Recommeded to run these files on a High-performance cluster for the parallel computing process.

Following this, run the MATLAB code named "NAFEMS_Problem_2.m" followed by the R code named "NAFEMS2.R"

Reference(s):

• NAFEMS Stochastic Working Group (2022). Stochastic Challenge Problem: Uncertain Knowlege. Stochastic Challenge Problems Website. URL: https://www.nafems.org/community/working-groups/stochastics/challenge_problem/
• A. Lye, A. Gray, M. de-Angelis, and S. Ferson (2023). Robust Probability Bounds Analysis for Failure Analysis under Lack of Data and Model Uncertainty. In Proceedings of the 5th International Conference on Uncertainty Quantication in Computational Sciences and Engineering, 1, 391-407. doi: 10.7712/120223.10345.19797

Author:

• Name: Adolphus Lye
• Contact: snrltsa@nus.edu.sg
• Affiliation: Singapore Nuclear Research and Safety Institute, National University of Singapore