Edge computing-based virtual measuring machine for process-parallel prediction of workpiece quality in metal cutting

Digital process twins enable in-line quality inspection and root cause analysis, thus sustainably optimizing the production ramp-up and quality control cycle. Current approaches reply on expensive computing units or platforms to perform material removal simulations followed by virtual metrology. This raises deployment costs and conflicts with industrial edge devices. In this paper, a novel analytical approach is proposed to build an edge computing-based digital twin of machining processes for the process-parallel prediction of workpiece tolerances by leveraging machine internal signals and contextualizing metadata sources along the digital process chain, i.e., the digital thread. This concept is validated with the results of a state-of-the-art digital process twin and conventional measurement data from a coordinate measuring machine.