Open-source and probabilistic materials models

This deliverable presents a comprehensive computational framework for advancing the probabilistic design and manufacturing of next generation airplanes. Leveraging machine learning algorithms, the project integrates into the simulation workflow, enhancing efficiency and effectiveness in aircraft design. By proposing a framework linking material properties to defects and material properties through machine learning surrogates, CAELESTIS paves the way for large-scale simulation campaigns within practical timeframes. This approach will offer faster and more precise results for optimizing both product and process along the manufacturing chain.  The deliverable includes open-source machine learning surrogates for probabilistic material law predictions, as well as an open-source simulation solution used to generate the data required for surrogate training. The proposed code will be made available through a public repository.