Data-Driven High-Fidelity Modeling


In daily industrial practice, Computational Engineering faces two critical issues: efficiency and credibility. Indeed, efficient strategies are needed to carry out computationally-demanding multiple queries of complex multi-physics and multi-disciplinary problems arising in parametric studies. Besides, the output of the model is expected to be credible, namely with guaranteed numerical accuracy and quantified and controlled uncertainty. In a data-driven approach, data is used to update the model and quantify uncertainty.

The group responds to these challenges adopting a comprehensive approach in the discipline of computational science and engineering, developing new mathematical models and numerical methods to produce high-fidelity solutions in a variety of complex interdisciplinary problems.

The group is also active in the development of open-source software and in the integration of innovative algorithms in existing open-source libraries.