UQLab is a general-purpose uncertainty quantification (UQ) software developed in MATLAB at the Chair of Risk, Safety and Uncertainty Quantification, ETH Zurich, Switzerland. It is designed to bring state-of-the-art UQ techniques and algorithms to a large audience.
UQLab is made of open-source scientific modules which are smoothly connected through UQLabCore, a content management system that allows for an easy use of the modules and user-written extensions. As of version V1.2, the modules cover:
- Uncertainty propagation through Monte Carlo sampling
- Metamodeling (e.g., polynomial chaos expansions, Kriging, support vector machines, low-rank tensor representations)
- Sensitivity analysis
- Reliability analysis (i.e., computation of rare event probabilities)
- Bayesian inversion analysis for model calibration
- …and more!
UQLab is not simply an umpteenth toolbox for UQ, but a framework: not only it offers you an extensive arsenal of built-in types of analyses and algorithms, but it also provides a powerful new way of developing and implementing your own ideas.
MATLAB® is a registered trademark of The Mathworks, Inc.