Automating Defect Detection in Laminate Manufacturing Process 

Manual quality inspection has reached the edge of human capabilities; that’s why we use AI.

— Klaus Lozie, Digital Operations Manager, Unilin Group

Robovision & Unilin Case Study

Real-time detection of rare and hard-to-see defects is not only a challenge, but it also creates an opportunity for manufacturers to raise the bar on their quality standards, increase efficiency for operators and improve production yields. The Unilin Group partnered with Robovision to build an AI-powered in-line quality control system that automates visual inspection for their product category – laminate flooring.

For one of the world’s largest flooring manufacturers, the Unilin Group, mostly known for its successful brand Quick-Step, uses Artificial Intelligence (AI) to ensure that they deliver the highest quality laminate flooring. They were looking for a solution partner, who can help them build an AI-powered in-line quality control system to aid their operators in detecting defects.  

Their challenge was threefold. First, the high assembly line speed (100 metres / minute) makes it difficult for even experienced operators to accurately spot all possible flaws. Second, the shortage of trained personnels or domain experts is a huge constraint for manual quality inspection. Third, as their laminate flooring is almost the same as real wood, it is not easy to tell if it is a defect or just a natural element of the print. 

As a quality-driven company, automating the quality control process with accurate defect detection can help them achieve a competitive edge. The Unilin Group approached Robovision – an award-winning leader in AI and computer-vision technology – to develop a scalable solution with their easy-to-use AI platform to automatically detect defects, across different laminate types and colours. Operators are alerted in real time when a defect occurs, so they can immediately remove the defective product out of production. They can also adjust production factors if necessary, in order to avoid producing more products with the same defect.

Thanks to the AI quality control system, the Unilin Group was able to achieve a higher production output and streamline the visual inspection process. 

The Challenge

The quest for continuous quality Improvement

Embracing innovation, the Unilin Group constantly strives to deliver the highest quality through emerging technology. As the laminate plates pass by at 100 metres per minute, it is very hard for operators to see or detect the small defects inline. In the event of an error, the operators have to check different laminate packs offline to ensure the quality of end products. To avoid double work and increase efficiency, continuous defect monitoring was necessary. 

Furthermore, manual quality inspection has reached the edge of human capabilities. As qualified operators are harder to find, full automation was considered to overcome talent shortage. 

The search for a cutting-edge solution began when the Unilin Group realised that AI will help reduce defective end products and increase efficiency for staff members. By removing time-consuming manual inspection, domain experts can focus on early intervention: analysing the defect and taking timely actions in the production process. As a result, they can increase production yields and streamline the inline quality control process with automation. 

Unilin-group-defect-detector-robovision

Figure 1: the Unilin defect detector

Manufacturing-Defect-Detection-with-AI-computer-vision-final-check

Figure 2: Collaborative Intelligence – human operative using AI 

The Solution

Solving the problem at production scale

AI-powered defect detection is considered to be the future of visual inspection. The use of AI-enabled systems combined with cameras allows manufacturers to deliver high quality with reduced time delay and defect costs. 

Together with the Unilin Group’s team, Robovision helped build an automated in-line quality control system. The system can detect even microscopic flaws in a high-speed production environment. An in-line camera scans critical zones of the laminate board, then the AI model processes the data and provides visual insights. If a defect is detected, the operator is alerted immediately to remove the defective laminate board out of production. Operators can now focus on other tasks while letting AI do its work. 

Within the Robovision AI Platform (RVAI), the Robovision team developed an AI model that is scalable across different product types. For the Unilin Group, the team can easily retrain the model on different laminate types and colours, with easy deployment across production lines and factories. 

Thanks to AI, monitoring defects of every single laminate plate is now feasible in such a complex and fast-moving production process.

— Dries Van Poucke, Production Engineer, Unilin Group

The Approach

Deep learning technique - a game changer

Detecting defects on laminate is much more challenging and subtle than for example detecting people in a picture. As laminate has almost the same natural features as real wood, it is hard to see if there is a defect or it is just an element of the print. Traditional computer vision would not be able to handle these subtle changes of the laminate dataset. 

Therefore we used a deep learning AI model that can now detect defects on different types and colours of laminate. 

Deep-learning-technique-for-hard-to-see-defects

Figure 3: Defect in a laminate board

Robovision-AI-platform-for-quality-control-systems

Figure 4: AI-powered Robovision Platform

The Technology

A powerful platform for AI-driven companies

The Robovision AI (RVAI) platform was provided as part of the total solution. Robovision first helped the Unilin Group select a carefully designed architecture to build the AI models leveraging NVIDIA GPU processing power, within the platform. Then, using the iterative approach, we were able to develop a highly accurate AI classification model that was tested on both production samples and difficult test sets. We then gave the Unilin Group’s experts training on the RVAI platform. 

The Unilin Group’s team can use the RVAI platform on-site to not only improve accuracy but also to scale what the model can do on their own. Without any AI knowledge or having a data science team, they can deploy and manage the lifecycle of their AI models flexibly and autonomously. Their experts can easily retrain the model on new types and new colours of laminate by themselves. 

Thanks to the Robovision AI platform, Robovision and the Unilin Group were able to join forces. Robovision’s AI expertise was combined with Unilin’s product insights and problem understanding to successfully implement an AI-powered in-line quality control system in production. 

The AI model automatically alerts operators whenever a defect occurs on the laminate board, this enables them to immediately remove it out of production.

— Pieter Blomme, Software Engineer, Robovision
unilin-robovision-continuous-monitoring-of-defects

Reaching Desirable Outcomes

AUTOMATE COMPLEX VISUAL TASKS

Detect subtle defects in a high-speed production environment.

CONTINUOUS MONITORING

Spot specific defects on every single laminate plate.

INCREASED EFFICIENCY

Help operators focus on other production tasks.

ACCURATE DEFECT DETECTION

Deep-learning models become smarter and autonomously learn all defects on different laminate types.

EASY TO ROLL OUT AI MODELS

Access to an easy-to-use platform with no AI knowledge required.

A HIGHLY SCALABLE SOLUTION

Scaling the solution across product types and production lines.

AI in combination with the Robovision AI platform can help your operators achieve the highest quality standards for your end customers.

— Julie Coorevits, Account Manager at Robovision

Join Unilin in the AI Revolution

Start the discussion with Robovision today

Contact the Team