AI as a digital operator: smarter collaboration on the production line
March 31, 2026 • 3 minutes reading
In a nutshell:
- The production of ultra-realistic laminate requires extreme precision
- Unilin Flooring developed AI solutions in-house with its production staff
- AI supports operators in production, without taking control
One of Unilin Flooring’s showpieces is the Capture collection: laminate with the look and feel of real wood. Its hyper-realistic appearance makes the production process particularly complex. The design and the pressed texture must align perfectly, while the boards move through the line at high speed.
For operator Samuel Vandenberghe, who has been working on the Lux18 line for two and a half years, the challenge is clear: “With the Capture laminate floor collection, it comes down to fractions of a millimetre. For us humans it’s not always possible to work with that degree of precision, especially at the speed the boards are fed through. Producing Capture without AI is simply not an option."
AI Finder and Argus: precision developed by and for flooring
That precision support was developed with ‘AI Finder’, a proprietary camera technology developed by Unilin Flooring. Before pressing, cameras analyse both the decor and the base panel. The system automatically identifies key points and aligns the layers pixel-perfect. Thanks to ‘Argus’, AI also supports operators after pressing. Argus is a deep learning camera system developed in-house by the Digital Operations Flooring team. Argus inspects each individual panel for defects, including those not visible to the naked eye. Samuel: “In the past, we carried out spot checks to detect recurring errors. Today, nothing escapes Argus’s spying eye. With the aid of AI as a digital operator, we are able to deliver even higher quality.”
From classic vision system to deep learning
Over the past few years, AI engineer Pieter Gobin has contributed to the development of Argus. He witnessed firsthand how Unilin evolved from classic image recognition to learning systems. “Unilin’s classic image recognition technology Univision, which was introduced in the early 2000s, already achieved an accuracy of 99% but that was the ceiling.”
Deep learning has pushed that boundary. “With AI, we are aiming for 99.9% or more. This translates into less downtime, higher output, and above all, greater confidence in quality,” says Pieter. “The number of rejected products has more than halved.” To train Argus, the team collected around 40,000 images. “We fed them to the model until it was able to classify the images correctly,” Pieter explains. “Today, Argus assesses each photo inline and identifies even the smallest deviations.”“With AI, we are aiming for 99.9% accuracy or more. This translates into less downtime, higher output, and above all, greater confidence in quality”
Innovation with the people on the floor
For Pieter and his team, one principle has always been central: technology must serve the people who use it. "AI must offer clear added value for every user. Argus is not a self-correcting system. We rely on the knowledge and experience of our operators.” Samuel confirms: "We are in control at all times. AI provides support but we still make all the decisions.”
What began as a solution for a single complex production line has grown into a global success. AI Finder has since been rolled out to other production sites, including in the United States. Global support is provided from Wielsbeke.
Are you inspired by innovation?
Read more
Unilin takes the circular economy to the next level with recycling line for MDF, HDF and laminate
Unilin has achieved a world first in Bazeilles, France: for the first time, MDF and HDF boards – the core of laminate floors – can be recycled on an industrial scale. . Thanks to a €20 million investment, the company is now able to make the wood fibres from these boards reusable for the production of new decorative panels and laminate floors.
AI at Unilin: assess what is there and make improvements
Six months ago a global AI hype took the world by storm. Unilin was in no hurry to climb the bandwagon because we have been investigating the potential benefits of AI for all our processes for several years now. Meanwhile some AI models have been integrated into the production process while others are currently being trained and tested. Still others were found to be too unreliable after a test phase. The key question is: when does AI provide us with added value?
.ashx?h=1080&iar=0&w=1920&rev=0fd2bc11c6c74d3b87419b103a1f47ee&hash=4E18904D53432B1FF9C75375B3DA54E1)
From waste to resource: how Unilin recycles laminate floors
What happens to your old laminate floor? At Unilin, the answer is clear: collect and recycle. Through the Recover programme, we collect old floors, which are then recycled into new MDF and HDF panels using a technological world first. “For years, recycling the panels seemed impossible,” says Geert Coudenys, R&D Director at Unilin Panels.