Volkswagen Using Computer Vision To Increase Production Efficiency
Manufacturing may indeed be an odd thing to get excited about, but it seems that the more you dig in, the more fascinating it is. In this day and age, with technology evolving and improving so rapidly, there’s enormous potential to improve manufacturing quality and efficiency.
We highlighted that last year in our tour of Tesla factories in Fremont, California. (See here, here, here, here, here, and here, for example. Also check out my summary of our interview with Tesla President Jerome Guillen.) We’ve also spent some time in the Volkswagen factory that will first produce the ID.3, as well as a few other electric vehicle models, but that was mostly a superficial visit and press photo op. Now, though, we have a little more insight into how Volkswagen is innovating in the manufacturing world.
Volkswagen, as the title notes, is using “Computer Vision” to increase manufacturing efficiency at its factories. What is Computer Vision? Here’s how Volkswagen summarizes it: “The process extracts information from optical data, such as the real environment at the plant, which it then evaluates using artificial intelligence (AI). The procedure is similar to the human capability of recognizing, processing and analyzing images. Volkswagen has been working with this technology for several years and is now intensifying its efforts.”
Does This Really Matter?
How much does a little bit of computer vision help to cut energy use and cut costs? Quite a lot, actually. “Expect cost reductions running into the double-digit million range by 2024,” said Gerd Walker, Head of Volkswagen Group Production.
Volkswagen expects its “Industrial Computer Vision” image recognition and processing technology to lead to a 30% productivity improvement in its production processes from 2016 to 2025. In other words, we’re about halfway through this evolution, which should lead to a decent estimate of the long-term productivity improvement.
Their productivity push is focused on finding solutions that can be implemented across Volkswagen Group factories, thus benefiting from economies of scale.
“The first two Computer Vision solutions from Porsche and Audi are currently being prepared for Group-wide rollout and connection to the Volkswagen Industrial Cloud,” the company notes.
By now, you’ve noticed the somewhat odd photos of a factory worker looking at her phone and some vehicle stickers. If you’re as clever as Sherlock Holmes, you’ve gleaned that this is somehow related to the Computer Vision and production efficiency Volkswagen is hyping. If you’re not entirely clear what these pictures are telling you, join the club. But Volkswagen is here to clarify the news:
“The first application, which is to be rolled out via the new Volkswagen Industrial Cloud throughout the Group next year, is currently being tested by Porsche in Leipzig. The application functions as follows: several labels are attached to each vehicle produced, for example with vehicle information or notes on airbags. Many of these labels contain country-specific information and are written in the customer’s language. The proper application of these labels is ensured by Computer Vision.
“At the Porsche plant in Leipzig, an employee on the production line now scans the vehicle identification number to ensure clear identification of the vehicle. Photos are taken of each label attached to the car. The app checks the images to ensure that the labels have the correct content and are written in the appropriate language on a real-time basis and provides the production line employee with feedback on whether everything is correct. This saves several minutes per vehicle. The app was developed jointly by Porsche, the Volkswagen Software Development Center in Dresden and the Smart.Production:Lab Wolfsburg.”
I’m not entirely clear how this saves several minutes per vehicle, but I assume that factory workers previously had to spend a lot of time verifying themselves whether these stickers were appropriately placed. Now, an app rapidly tells them and they move onto the next one.
Note that minutes per vehicle is a big deal when it comes to a company that produces 10 million or so vehicles a year. Frugal Moogal explained this kind of thing well for CleanTechnica a few days ago, and it was also highlighted for us on a tour of Tesla’s seat factory last year (see links at top) in which lead engineers at the factory pointed out innovations that led to a few seconds saved per car. Combining that highlight with an earlier emphasis from Elon Musk on the need to cut pennies off the cost of producing each Model 3, CleanTechnica Digital Design Leader Chanan Bos created this fun graphic:
Volkswagen’s ability to now save minutes per car produced is no small matter. It’s a significant improvement in manufacturing efficiency.
Cameras + Machine Learning = More Precise Production
Many Germans pride themselves on the exceptional engineering and build quality of German vehicles. People around the world hum the same tune. Volkswagen Group is committed to going further, though.
“Another solution currently being prepared for use throughout the Group comes from Ingolstadt, where Audi uses it for quality testing at the press shop. Cameras combined with software based on machine learning detect the finest cracks and defects in components.”
Hopefully that newfound ability won’t lead to crippling perfectionism.
How Big Is The Team?
The crew of Computer Vision experts at Volkswagen Group isn’t that small of a group either. It consists of ~60 Computer Vision experts. They are constantly looking for ways the company can benefit from Computer Vision to improve their manufacturing quality or efficiency, or both.
Furthermore, they are going beyond the factories. “In addition to the use of the technology in production, Volkswagen plans applications along the entire value stream, for example in sales and after-sales. For development work on the optical procedure, Volkswagen is recruiting experts for this area in Berlin, Dresden, Munich and Wolfsburg. In addition, the Group continues to build up its skills in the fields of camera technology, machine learning and the operation of Computer Vision solutions.”
No doubt, as Volkswagen transitions a factory that just built its last fossil fuel vehicle after 116 years into a fully electric vehicle factory, and then completed the transition for several other factories around the globe, Volkswagen is looking for every opportunity it can to bring manufacturing up to a 2020s level to utilizes the best tech, whether it be basic tech like a smartphone and an app or new manufacturing robots for the factory floor. Stay tuned. As we get more information — and perhaps some factory tours (hint, hint) — we’ll be sure to share those insights with you.
Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.
Latest CleanTechnica.TV Video
CleanTechnica uses affiliate links. See our policy here.
