How is AI being used in the wood products industry?

AI in the wood products industry is used mostly at the scanner and the saw. It detects knots and cracks, grades boards and lamellas, chooses cutting patterns and adjusts machines as they run. The figures attached to it are harder to use than the list of uses. A gain quoted as less waste, one quoted as more yield and one quoted as more value can describe the same result. What decides how large it looks is the baseline it was measured against. That, more than the list of uses, decides what a real project will return.

What the usual answer says

The trade coverage that ranks for this question describes a fast-moving field. A 2026 feature in Wood & Panel Europe on smart woodworking describes scanners that map internal defects before cutting: "These systems automatically calculate the most efficient sawing paths, successfully reducing material waste by over 30%". It gives no source or baseline. A producer's own account of AI at its planer mills, from Domtar, quotes a quality manager: "There's a lot less waste". It gives no figure.

The list of uses in these articles is broadly right. The numbers, where there are any, are where a buyer should slow down.

One result, three ways to say it

A 2024 study in Sensors of a two-sided defect detection and cutting system shows why. The authors report that their cutting strategy "yields a notable 12.3% increase in the volume yield of sawn timber compared to present production". The full text gives the detail: on 100 pieces, yield rose from 69.2% with conventional fixed-length cutting to 81.5%.

Read as waste, the same test cut the share of wood lost from 30.8% to 18.5%, a reduction of about 40%. Read as relative yield, it is a gain of about 18%. So "12.3%", "about 18%" and "about 40% less waste" are one result. A headline of "over 30% less waste" is not out of line with it. The weak point in both is the baseline: fixed-length cutting is the simplest method a line can use, and a plant already running an optimiser starts much higher.

From the machine to the accounts

Even a well-described figure needs translating before it reaches the accounts. A yield gain applies only to the wood that passes that machine, and only on products where the defect it detects was the limit. A gain of 12.3 points at one saw becomes a smaller percentage of the whole business once the wood that never passes that saw is counted. The accounts will show that smaller figure.

Where prices change the answer

Value depends on the price list as well as the cut. A 2017 study in Annals of Forest Science by Rais and colleagues, three of whom worked for the maker of the CT scanner tested, simulated sawing 36 logs at their best rotation. It found value gains between 4% and 20%, depending on the price gap between strength grades. It is small, simulated and older than five years, but its point stands: a value gain reported under one price list is not a value gain under another.

Beyond the machine

Purchasing, stock, scheduling and pricing depend on records in the ERP and production logs, not images. Vendors sell AI for those jobs, and the top results mention forecasting and stock counting. What we could not find is a published measurement of a gain in them, which is a reason to ask for one before buying.

How to read a claimed gain

Ask what the figure measures: waste, yield or value, and relative or in points. Ask what it was measured against, because a gain over fixed-length cutting says little to a mill that already optimises. Ask how many boards or logs were tested, from which operation, and at what prices. Then convert the claim into the units your own accounts use before comparing it with anything else.

When it doesn't apply

Some AI uses in wood products are not sold on yield at all. Safety monitoring, which detects people or material in danger zones, and labour-saving automation are judged on incidents and hours. A vendor who offers a trial on your own boards, against your current method, is offering the most relevant measurement there is.

Quarri for wood products is built around how a wood products plant runs, where the order book meets real capacity.

Quarri is an AI-native data platform for the timber supply chain. It connects buying, production, sales and inventory for forest management, sawmill, wood products and pulp, paper and packaging operators.