68th Annual SVC Technical Conference • May 19 - 22, 2025

Education Program • May 17 - 22 | Technology Exhibit • May 20 - 21

Digital Transformation through Artificial Intelligence, Machine Learning, Simulation, and Data Science in the Thin Film Industry

This session covers all topics in which novel digital technologies play an important role. These include, without limitation, physics and chemistry simulations, advanced data science techniques, and approaches that rely on subsets of artificial intelligence, such as machine learning. It brings together experts in simulation and artificial intelligence and provides an ideal platform to discuss the benefits of the digital transformation of industrial deposition processes from the perspective of various technology fields. The session welcomes perspectives from academic experts as well as stakeholders from the entire vacuum coating supply chain — OEMs, coating centers, providers of coater components and monitoring tools, and providers of digital services and simulation software.

The motivation behind this session is the fact that industrial deposition processes are under strong competitive pressure, as better productivity is always demanded with higher precision and increasing complexity of coating products. This increased complexity requires optimized coating processes, model-based process control, and a comprehensive view and understanding of the entire process chain. Therefore, a digital transformation, which will be one of the key drivers in the future for industrial deposition processes, is needed.

The digital transformation includes the systematic collection of data generated in different processes and the representation of the coating processes through real-time capable digital twins.

Even today, simulation and digital twin models are well-established tools for predicting and optimizing deposition processes. It is possible to use physical and/or chemical models to predict the behavior of the process with very little a priori knowledge.

Another approach to predicting processes is the use of generated data and components of artificial intelligence, such as machine learning, deep learning, or grey-box models. In this context, data acquisition, storage, and accessibility become increasingly important. Artificial intelligence is already deployed in areas such as image recognition, predictive maintenance, and process control.

Digital Transformation through Artificial Intelligence, Machine Learning, Simulation, and Data Science in the Thin Film Industry TAC Co-Chairs: Holger Gerdes, Fraunhofer-IST, holger.gerdes@ist.fraunhofer.de; Adam Obrusnik, PlasmaSolve s.r.o., obrusnik@plasmasolve.com