Uncertainty Quantification Tutorial using Jupyter Notebooks

By Ilias Bilionis

Mechanical Engineering, Purdue University, West Lafayette, IN

Published on


Increasing modeling detail is not necessarily correlated with increasing predictive ability. Setting modeling and numerical discretization errors aside, the more detailed a model gets, the larger the number of parameters required to accurately specify its initial/boundary conditions, constitutive laws, external forcing, object geometries, etc. To be predictive, we need to quantify this uncertainty by combining our prior physical knowledge with noisy experimental data obtained from various heterogeneous sources. Once we have quantified this uncertainty, all we need to do is propagate it through the model and obtain predictive error bars for any quantity of interest. What kinds of uncertainty do we encounter in physical models? How is uncertainty described mathematically? What can go wrong if uncertainty is ignored?

This tutorial includes files to be uploaded into your nanoHUB Jupyter notebook session (go to https://nanohub.org/tools/jupyter).  4 documents are included in the default zip file. The pdf file includes the instructions on what to do with the other 3 files in this package, which include data, an image, and an ipython notebook.  Open the "supporting documents" tab for this resource if you'd like to see just the pdf file.

Cite this work

Researchers should cite this work as follows:

  • Ilias Bilionis (2018), "Uncertainty Quantification Tutorial using Jupyter Notebooks," http://nanohub.org/resources/28946.

    BibTex | EndNote


Tanya Faltens

Network for Computational Nanotechnology, Purdue University, West Lafayette, IN