Traditional stoneware manufacturer adds AI quality control to manual inspection.
Zahna Fliesen has been pressing tiles in Saxony-Anhalt since 1891. Most of the inspection still happens manually under shop-floor lighting, which is great for craft control and tough on consistency. Surface flaws and color shifts can slip through, and training new inspectors takes weeks.
Enao runs on an iPhone mounted above one of the inspection tables. The model flags chipping, hairline cracks, and out-of-spec coloring as tiles pass underneath. Inspectors confirm or correct each call on the same screen, and the model adjusts to new patterns within a shift.
“We didn't expect a phone to keep up with our line. It does, and it gets better every week.”
Zahna Fliesen has been pressing tiles in Saxony-Anhalt since 1891. The factory still runs lines that have not changed shape in decades, and quality control still happens by eye. That craft is part of what makes the brand. It is also a long, careful, fatiguing job.
The team had looked at machine vision before. Every quote came back with a six-figure price tag and a six-month timeline. Tile patterns also change often, which means the system would need recalibration each time. That was a non-starter for a factory of this size.
Setup took an afternoon. By the end of week one the model agreed with senior inspectors most of the time on patterns it had already seen. By week four it was pulling its weight on new patterns too, after a few rounds of operator feedback. No new wiring, no new servers, no extra desk in the QA office.
We use cookies to understand how visitors use our site so we can improve it. Analytics only run if you accept. You can change your choice anytime in the footer. Privacy Policy.