The group “Data-driven Modelling and Artificial Intelligence” at the IAM-MMS specialises in developing, providing, and integrating artificial intelligence (AI) methods for applications in materials science and mechanics. In cooperation with the other groups of the IAM-MMS and within the framework of collaborations at the KIT and external partners, we provide solutions for the characterisation, analysis, and synthesis of materials, especially microstructures, with material behaviour ranging from fracture mechanics to biological processes. In addition to experimental data, e. g., from imaging processes, many applications also use simulation methods such as the finite element method (FEM) or finite volume method (FVM) to generate additional data of physical systems. For this purpose, physics-based and data-driven methods are combined in intelligent elements and solvers, e. g., to enable a dimensional reduction and an efficient computation. The AI and ML methods include supervised, unsupervised, reinforcement, and active learning, ranging from principal component analysis to deep neural networks, tailored for regression, classification, and clustering tasks. In combination with research data management, optimisation algorithms allow the implementation of design-of-experiments processes, e. g., with Bayesian optimisation. Our interdisciplinary scope includes both data-science and engineering expertise, which facilitates the development of explainable and physics-informed AI methods. A data platform and efficient research data management are essential for effective implementation. Therefore, the group develops the open-source Python library CIDS (Computational Intelligence and Data Science) for applications of AI in materials science and mechanics. This library also forms the basis for the AI pillar of the research data platform Kadi4Mat: KadiAI. This direct integration into Kadi efficiently couples data and physics and enables best-practice workflows that provide AI solutions for physical problems.

• Explainable and physics-informed machine learning models for experimental studies and simulations of materials

• Coupling data-driven models with physics-based simulation software

• Integration and interaction of AI methods and research data management for active learning and automated experimental design