Please join us on Wednesday, April 22 for our next HEMI Seminar.
Speaker: Chris Eberl
Deputy Director,
Fraunhofer Institute for Mechanics of Materials IWM
Title: Data driven materials development and lifetime prediction – challenges and opportunities for accelerating and advancing Materials Science and Engineering
Abstract: The advancement of available AI models allows to tackle many different challenges in Materials Science and Engineering. In this talk, actual applications of different AI models shall be discussed, covering a multitude of scientific workflows and tasks in MSE. The following examples will be discussed: (1) materials selection via random forest models, (2) classification and segmentation for microstructure analysis as well as damage identification and quantification via deep learning, (3) structured data from high-throughput testing for fatigue lifetime prediction via graph neural networks and explainable AI approaches, (4) data and information extraction from publications via Large Language Models and (5) semantic extraction and ontology learning for MSE via agentic AI.
While AI accelerates and enriches our understanding of materials, reliability and uncertainties need to be critically reviewed. Therefore, training data, training methodologies as well as model selection are all relevant ingrediencies and will be discussed.
Bio: Christoph Eberl is deputy director of the Fraunhofer Institute for Mechanics of Materials IWM in Freiburg and head of the Chair of Micro- and Materials Mechanics at the University of Freiburg. He initiated the Fraunhofer Cluster of Excellence for Programmable Materials and serves as its scientific coordinator. He is also a thought leader in the digitalization of materials research, and he believes that the digital transformation of materials research is the key to international competitiveness in science and industry, as well as to addressing global challenges such as climate change, resource scarcity, and energy supply.
