Skip to main content
Machine Learning

In industry, product design and subsequent process planning is becoming increasingly complex. The reasons are increased customer requirements and a trend towards smaller batch sizes. Especially, the dimensional accuracy is crucial for most machined parts, since it affects directly the functionality. To support process planning, the industry often uses process simulations. Nevertheless, manual adjustments are still necessary to compensate for process- or machine-specific deviations such as tool stiffness or machine dynamics.  
Under the direction of Prof. Dr. Eirini Ntoutsi from L3S and Dr.-Ing. Marc-André Dittrich from the Institute of Production Engineering and Machine Tools at Leibniz Universität Hannover, the DFG-funded project "Hephaestus: Machine Learning Strategies for Adaptive Process Planning in 5-axis Milling" will investigate, among other things, how shape errors in milling processes can be compensated on the basis of a process-parallel material removal simulation and sophisticated machine learning (ML) strategies. The project starts in October 2020 and will last for 2.5 years.