- AI-assisted testing instead of visual checks: The new system utilizes fully automated final inspections of closed-die forgings, thus replacing conventional manual checks
- Comprehensive quality assurance: Automated robot handling and component inspections within seconds provide objective quality control for every single component
- Test system scalability: The use of synthetic training data enables the fast, cost-efficient integration of new product variants and reduces personnel and process costs on a sustainable basis
“Our mission is to provide our customers with components manufactured with uncompromising precision. That’s why with the new automated final inspection system, we are taking the next logical step: Not only do we guarantee superior part quality, but we also increase our line’s productivity thanks to the high degree of automation. This collaborative partnership allows us to integrate state-of-the-art testing technology into the production process and thus solidify our standards on a lasting basis,” says Thomas Baur, senior production manager at Hirschvogel.
“We integrate AI directly into the harsh environment of a forge shop. The cooperation with Hirschvogel is an excellent opportunity to work together to digitally advance the solid steel forming process. It’s great to have a partner that is so consistent when it comes to innovation. The new system enables us to speed up the final inspection significantly and ensure maximum process stability in the line itself” says Axel Rossbach, forging technology expert at SMS group.
The centerpiece of this solution is a platform that efficiently trains analysis models using synthetic data. Instead of collecting real defect images on physical parts over the course of weeks, the system uses CAD data or 3D models. Material properties and surface textures are digitally simulated so precisely that they accurately reflect actual conditions. An automated process generates photorealistic image data and virtually applies various types of defects to the components. The result: a ready-to-use analysis model in just six to eight hours.
The inspection process at Hirschvogel includes a fully automated feeding system: a robot removes the forgings directly from the transport crate. High-resolution camera sensors capture the components while the software analyzes them in fractions of a second. The system reliably identifies geometry deviations and surface defects and allocates them precisely to the digital defect catalog. This objective AI evaluation minimizes the error rate in classification, which means that pseudo-scrapping—in other words, the erroneous disposal of good parts—can be almost completely avoided.
The benefits for solid metal forming are measurable: High-level automation reduces dependence on personnel, cuts labor costs, and eliminates human error sources. The high scalability also allows the system to be adapted quickly to new product variants.