Panneau métallique dans la cabine de peinture en poudre

How artificial intelligence could improve powder coating

Applying powder means continuously adjusting about ten variables: flow rate, voltage, distance, speed, air humidity. A good applicator does it by eye and by experience. Recently, software has been claiming to do the same thing from data. Here is what really exists, and what is still a promise.

Measure before curing

It all starts with measurement. Most often, film thickness is measured after curing, when it is too late to correct. The Swiss company coatmaster sells a system that measures thickness without contact, over the whole part, before curing. At Wagner, the Flowsense sensor measures in grams per minute the powder coming out of each gun and flags nozzle wear.

Software that adjusts the guns

With this data, software can suggest settings. That is what coatingAI does, a young Swiss company whose tool combines thickness measurements, 3D scanning of parts and ambient sensors to adjust the guns and their movement axes, according to a Products Finishing report. AkzoNobel partnered with it in December 2023. A first named customer, the British company Radii Planet Group, reports uniformity improved by 9.3% and powder loss reduced by more than 6%.

Steel panel powder coated in blue on its rack
An even film over the whole surface: that is what these tools aim to guarantee.

What remains to be proven

The powder savings announced by suppliers range from 16 to 30%. We found no result verified by an independent third party: all the figures come from the vendors or their customers. Research is also moving forward. In 2025, a Croatian team published in Applied Sciences a model that predicts film thickness from the gun's voltage and current, in the laboratory. The PaintExpo 2026 trade show also made artificial intelligence one of its themes.

And in a shop like ours?

These systems target automatic lines, where the same part goes by thousands of times. In a job shop, the parts change from one order to the next, and it is still the applicator's hand that does the work. We do, however, see three realistic uses in the medium term: thickness measurement before curing, settings suggested according to the shape of the part, and automatic tracking of oven parameters. Artificial intelligence will not replace metal preparation, but it could make every pass of the gun more consistent.

In summary

  • Thickness measurement before curing already exists and makes correction possible.
  • Software adjusts the guns from this data, on automatic lines.
  • The announced savings, from 16 to 30%, are not independently verified.
  • In a job shop, the tool will help the applicator more than it will replace them.

Free quote

Got a part to coat? Let’s talk.

Describe your parts in two minutes. We reply within 24 to 72 hours, on business days.