UQLab is a Matlab-based software framework designed to bring state-of-the art uncertainty quantification (UQ) techniques and algorithms to a large audience.

 

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. 

 

The project originated in 2013, when Prof. Bruno Sudret founded the Chair of Risk, Safety and Uncertainty Quantification at ETH Zurich, and decided to gather the results of a decade of his research into a single software tool. As of today, UQLab provides the software backbone of the Chair's research, allowing for seamless integration between the many research fields engaged by its members, e.g. metamodeling (polynomial chaos expansions, Gaussian process modelling (Kriging), low-rank tensor approximations), rare event estimation (structural reliability), global sensitivity analysis, Bayesian techniques for inverse problems, reliability-based design optimization, etc.

 

After more than two years of development we decided to open the platform to other research institutions, in an effort to increase the awareness of the scientific community on the fundamental aspects of uncertainty quantification. The first closed beta version was online from July 1st, 2015 until April 28th, 2017, allowing the Development team to collect valuable feedback from the users.

​

UQLab V1.0 was released in April 28th, 2017. With this release, the code was separated between UQLabCore, a content management system that seamlessly connects the open source UQLab Modules to carry out uncertainty propagation through Monte Carlo sampling, sensitivity analysis, reliability analysis (computation of rare event probabilities), build surrogate models (polynomial chaos expansions, Kriging, low-rank tensor approximations, etc.) and more.

​

As of September 2019, UQLab has kept growing both as a project and as a software, counting more and more features (see the development milestones ➜ ) and a growing user community (visit the UQWorld ➜ community website).