How to improve quality control and timber measurement - practical examples

Quality control of wood is a challenge

Quality control and, where applicable, accurate dimensional measurement are very demanding disciplines in the case of wood materials. Each piece is unique, so defects or deviations are difficult to determine from a reference product. 

Do you want to minimize the risk of a defective product being placed on the market or passed on to the next production process? You have two options - either improve the inspection process with human operators or fully automate quality inspection.

When the inspector is the one who manufactures

We'll look at both of these options, but first let's look at common practice in many manufacturing companies. Creating a separate position of quality inspector is challenging (financially and, given the labour market situation, also in terms of personnel). Therefore, such jobs often do not even exist. 

In practice, quality control is often the responsibility of an employee who is also in charge of production. A spot check is then carried out once in a while by the manager. When a problem arises, the manufacturer is able to find the culprit internally - the stacked products are marked with the name of the employee who was responsible for the quality. However, because the defective pieces have already reached the customer, this does not solve the core problem. The company has to deal with the complaint and the associated problems.

1) Quality inspector as a job position

This option is problematic in large-scale production because it requires the involvement of a larger number of workers. With this, the rate of inconsistent results due to subjective evaluation increases. 

For smaller production runs, you can leave the inspection process in the hands of an experienced inspector. However, you have to take into account the time required (even tens of seconds per piece) if he or she is to work meticulously. Increased pressure for speed would probably lead to frequent overlooking of defects.

But with a quality controller, the risk of putting in a defective piece doesn't go away, it's just minimized. Problems arise especially when more than one person is checking (at the same time, or on different shifts) - variations and noticeable variations in quality occur.

2) Automation of quality control and measurement

In the case of large-scale production, if every piece is to be checked honestly, it is essentially necessary to address inspection and measurement through an automated solution. Operators would only be able to carry out random checks, with the associated high risk of defective piece throughput.

However, even for medium and small companies, automation is worthwhile. Systems are becoming increasingly affordable in terms of the investment required, thanks in part to the rise of collaborative robots. They make robot workplace integration easy even in plants that have not yet taken any steps towards industrial automation. On the contrary, it represents a way to break into automation and robotics and gradually deploy it in other production processes.

The system detects defects with the help of AI

Due to the variety of materials and the uniqueness of each defect, wood inspection has long been problematic. At Kinali, we realized that comparing with reference products did not yield satisfactory results. 

That's why we developed WoodenAI Opti control system, which uses artificial intelligence to find defects. In addition, it can measure wooden products or raw wood from all sides, with an accuracy of hundredths of millimetres. At this point, it can detect the following defects with a high degree of reliability:

• cracks, splits, knots and broken corners, 
• areas of mould, wood staining or rot, 
• poorly machined areas or saw burns, 
• colour shades outside the specified range, 
• foreign objects such as nails or glue residues.

Automation in various areas of woodworking

A system for automated wood quality control has its place in many areas of wood production. For example, it is suitable for checking

• measuring tools - for sorting out unsuitable material and checking accuracy
• wood flooring - for quality grading or mosaic wood selection
• doors and windows - for checking the dimensional accuracy of parts before assembly
• mouldings and profiles - for discarding pieces with defects or unsatisfactory appearance
• furniture - to detect defects in units and individual components
• toys - to achieve high quality premium products
• sports equipment - to increase reliability and strength 
• wooden decorations - to improve the quality of premium quality products

Deployment of the system in the production of folding metres

We have successfully integrated the aforementioned WoodenAI Opti control system in Kinali into an operation that focuses on the production of folding meters. The quality and accuracy requirements here were high, as the meters fall under Class III accuracy and are supervised by the metrology authority.

We met the demanding requirement to measure as accurately as possible (to within hundredths of a mm) the 300 x 20 mm slats. The robotic workstation measures the slats from all sides at a speed of 700 pieces per minute. Pieces with defects or deviations are then automatically sorted into the appropriate NOC basket.

Are you interested in how to automate the inspection of wood products or other materials in your plant? Contact us. We would be happy to meet you in our showroom in Brno, where you can also see the robotic workstation in operation.