How BMW uses artificial intelligence to make car assembly more efficient

Artificial intelligence is having a huge impact on the automotive industry.

Revenue from sales of self-driving cars is expected to reach $70 billion by 2033, according to Future Market Insights. But AI self-driving cars aren’t the only change – AI technology has already been integrated into car production.

As part of this industry-wide trend, the BMW Group is now shifting gears to rely more on artificial intelligence to create a leaner and more efficient manufacturing process.

Over the past few years, BMW has She upgrades her Spartanburg, South Carolina factory to include new AI capabilities. The plant spans over 8 million square feet and produces approximately 60% of all BMW vehicles sold in the United States which amounts to more than 1,500 vehicles produced daily.

At The Body Shop, robots weld between 300 and 400 metal studs into the frame of each SUV. That’s about half a million studs a day that are applied by machines and now run by AI.

On top of that, AI technology checks to make sure each bolt is precisely positioned, according to BMW Group Principal Curtis Tingle. If a stud is out of place, the system tells the robots to correct it. No human intervention needed.

“It’s a complete closed loop,” Tingle told CNBC. “[AI] It removes human thinking, human manual intervention, right out of the equation.”

Tingle said the new technology has greatly improved efficiency. “We are achieving five times more than we thought possible even before, with what AI is achieving now.”

According to Tingle, the AI ​​nail correction laser has already saved the company more than $1 million annually. He said the new technology allowed BMW to remove six workers from the line.

BMW told CNBC that the AI ​​technology is patent pending and developed within the Spartanburg plant.

On the factory floor, Camille Roberts, Project Head of IT at BMW Group, explains that the new artificial intelligence software is helping speed up the carmaker’s existing inspection process.

As the SUV moves down the line, 26 different cameras throughout the grounds take pictures. That’s when, according to Roberts, “the AI ​​kicks in, identifies problems and reports them for a human to fix,” thus preventing the imperfect car from being shipped.

Roberts told CNBC that prior to the new AI upgrade, human workers could not inspect every vehicle as often as they can now, adding, “It’s not really possible to inspect every vehicle. … Production numbers are not going to meet global demand.”

There is still room to run for BMW’s AI technology, said Oliver Bilstein, BMW Group Vice President for Logistics and Production Control.

Workers at the plant wear what Bilstein calls factory scanners that take high-resolution measurements and images of every centimeter of the plant.

These images are used to build a 3D “digital twin” of the plant, Bilstein said, allowing BMW to instantly make adjustments and understand how they affect production before it implements a change in the real world. BMW plant planners around the world can access these detailed plans online.

With the help of the new AI software, Bilstein said, the scanning process now takes days instead of months.

Ultimately, this kind of AI technology will be able to learn, on its own, how to discover and recommend new ways to make the BMW Group’s automated assembly line more efficient, he said.