Self-optimizing process planning of multi-step polishing processes

Abstract Self-optimizing process planning is an essential approach for finding optimum parameters and reducing ramp-up times in machining processes. For this purpose, polishing presented as application example. In conventional processes, the are selected according to operator’s expertise order achieve a high-quality surface final production step. By implementing machine learning (ML) models planning, correlation between parameter measured quality generated. The of knowledge automates selection optimal computer-aided manufacturing (CAM) enables continuous adaptation NC-code changing conditions. Applying ML-model, prediction accuracy 83% will adapt target roughness 0.2 μm. sample efficiency shown by decrease root mean square error from 0.1–0.28 0.02–0.07 μm with additional iterations.