Artificial intelligence (AI) is transforming materials design and discovery by driving innovations in hardware architectures, software solutions for materials modelling and digital infrastructures collating data from computational and experimental characterisation and processing them using data science to enable learning and generate new knowledge. This session will explore how the design, optimization, and performance of computational tools are affected by this transformation on the path of enabling us to tackle more complex challenges with greater efficiency and accuracy.
Topics will include AI-augmented software frameworks that improve computational scalability, reduce costs, and expand predictive capabilities in materials modelling. Emphasis will be placed on the interface between AI and traditional modelling techniques, showcasing the role of ML methods in enhancing the accuracy and applicability of traditional simulations. Also, enriching experimental data sets with AI-generated data sets will be discussed.
The session will cover advancements in digital research infrastructures (DRI), such as high-performance computing (HPC), cloud platforms, and open-source tools. Discussions will address challenges with model interpretability, reproducibility and the use of heterogeneous architectures, as well as the adoption of collaborative platforms to accelerate innovation.
This session invites researchers, developers, and practitioners to share insights on their experience and understanding of AI's impact on modelling and simulation, emphasising its potential to drive further scientific discovery and technological progress.
Computational Chemistry: An Industrial Perspective
by Misbah Sarwar (Johnson Matthey, UK)
Measure what’s easy to measure, calculate what’s easy to calculate, use AI to fill in gaps
by Felix Hanke (Biovia R&D, Dassault Systems UK Limited , UK)
Digitalization of materials processing: the case of battery manufacturing
by Alejandro Franco (Université de Picardie Jules Verne, FR)
