TopicM: Modelling and Simulation
The focus of this symposium lies on data-driven methods that enable the design of new materials for desired properties, and on optimal control and planning methods for the corresponding processes. Such methods are typically based on the inversion of the process-structure-properties chain. Recent work shows that mappings between the individual elements of this chain can efficiently be established via data-driven methods that integrate domain knowledge, such as physical relations. Challenges in the modeling process are for example finding appropriate microstructure representations, accurate and robust machine learning approaches, and methods to efficiently generate data in the space of interest. The symposium provides a platform to discuss latest approaches tackling these challenges. It also aims at bringing together the communities of materials science, of data-driven modeling and machine learning, specifically the communities working on data-driven materials design and process design.