The business area for numerical data-driven prediction focuses on the handling of complex data from physical-technical applications, in particular in order to meet the challenges of industry 4.0. We combine mathematics, machine learning, and engineering knowledge to develop robust, scalable and domain-adapted data analysis concepts and methods. Applications can be found in the virtual product development based on CAE, in condition monitoring including predictive maintenance or in the realization of digital twins.
Our research is based on mathematical principles and aims in particular to integrate and use existing application knowledge. The scientific focus is on numerical methods for high-dimensional problems and the development of domain-specific data representations and similarity measures. We work on the efficient processing and analysis of large and complex data sets, for example time series or numerical simulation results, quantify uncertainties, provide highly developed tools for robust design or contribute to transfer learning.
We know that machine learning research in industry 4.0 is multidisciplinary. Therefore, we form a team that combines expertise in mathematics, computer science, engineering and physics.