Automated quality assurance
AI-supported testing of tool blades
Together with VisCheck, Hufschmied Zerspanungssysteme relies on AI-supported image processing to automate the quality control of tool blades. By using a collaborative robot, the inspection runs around the clock, while a digital twin saves all measurement results.
Together with VisCheck, an expert in intelligent image processing, Hufschmied Zerspanungssysteme has introduced a solution for automated quality control in the production of tool blades. Thanks to the AI-based OPDRA Quality solution, Hufschmied was able to significantly reduce the time and effort required for the quality inspection of its tool blades. The entire measurement of the tools and their visual inspection are now carried out fully automatically with the help of a collaborative robot. This enables the process to run around the clock. All measurement results are also stored in a digital twin. Each tool blade is provided with a data matrix code that enables the test results to be clearly assigned to the respective blade.
In addition to milling tools, Hufschmied also manufactures cold-cut blades at its plant in Winterlingen on the Swabian Alb, which are required in the automotive industry for airbag trimming. These extremely sensitive 20 × 5 mm blades undergo an optical inspection: an image processing system with machine learning software inspects the blade tip for potential damage in a grid of just 50 × 50 µm. The overall geometry with all lengths and angles is then recorded. To fully automate quality assurance, a collaborative robot removes the blades from a rack with more than 150 slots and performs the optical measurements overnight. The images and measured values generated during the inspection process are then stored in the cloud.
"With high-precision products such as Hufschmied's tools, the effort required for quality assurance sometimes exceeds that required for production. With our intelligent, adaptive image processing systems, we increase the efficiency of inspection processes and make them fully automatable," says Lutz Schaufuss, member of the VisCheck management board.









