How BMW uses AI to make vehicle assembly more efficient

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

According to Future Market Insights, self-driving car sales are expected to top $70 billion by 2033. But AI-powered self-driving cars aren’t the only change — AI technology is already being baked into vehicle production.

Related investment news

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

In recent years, BMW Upgraded its Spartanburg, South Carolina, factory to incorporate new artificial intelligence capabilities. The plant spans more than 8 million square feet and produces about 60 percent of BMW vehicles sold in the US, or more than 1,500 a day.

In the body shop, robots weld 300 to 400 metal studs to each SUV’s frame. About 500,000 studs are applied by machines every day and are now managed by artificial intelligence.

BMW group manager Curtis Tingle said that next, artificial intelligence technology checks to ensure that each stud is placed precisely. If a stud is misplaced, the system tells the robot to correct it. No human intervention is required.

“It’s a complete closed loop,” Tingle told CNBC. “(Artificial intelligence) directly eliminates human thinking and human intervention.”

Tingle said the new technology has dramatically improved efficiency. “With what AI is now achieving, we’re achieving five times more than we thought possible before.”

According to Tingle, AI stud correction lasers have already saved the company more than $1 million per year. The new technology has enabled BMW to remove six workers from the production line and redeploy them to other positions at the plant, he said.

BMW told CNBC that the patent-pending AI technology was developed at its Spartanburg plant.

On the factory floor, Camille Roberts, head of BMW Group IT projects, explained that new artificial intelligence software is helping to speed up the automaker’s existing inspection process.

26 different cameras throughout the floor snap pictures as the SUV drives down the production line. That’s when “artificial intelligence comes into play, identifying issues and flagging them for humans to fix,” preventing imperfect vehicles from being shipped, Roberts said.

Roberts told CNBC that before the new AI upgrade, human crews would not be able to inspect every car as they do now, adding, “It’s impossible for a human to inspect every car … production numbers won’t meet global demand.”

Oliver Bierstein, vice president of logistics and production control at BMW Group, said BMW’s artificial intelligence technology still has room to play.

Workers at the factory wear what Bilstein calls a factory scanner device that can measure and obtain high-resolution images of every centimeter of the factory.

These images are used to build a 3D “digital twin” of the factory, allowing BMW to make an immediate adjustment and see how it will affect production before implementing the change in the real world, Bilstein said. BMW plant planners around the world can access these detailed plans online.

With the help of new artificial intelligence software, the scanning process now takes days rather than months, Bilstein said.

Ultimately, the AI ​​technology will be able to learn on its own how to spot and recommend new ways to make the BMW Group’s automated assembly lines more efficient, he said.