Specifically, the project involves the analysis of spot welds in the bodywork. A delicate process that must be performed with the utmost care to avoid endangering the conditions of both the vehicle and the occupants. The development and testing of the technology took place in Neckarsulm, giving the expected results. Without going into too much detail, the company highlights its ability to examine about 1.5 million spot welds on 300 vehicles per shift.
Without AI, employees had to manually inspect about 5,000 points with ultrasound devices. The system will boost efficiency, and, as Audi claims, the analysis will be more accurate.
The new AI system promises to reduce the workload of the employees, who can focus on spotting any errors. But not everyone is happy about this. Many workers fear that they will lose their jobs to the machines, as the Manufacturer may not need so many human operators. The AI system has already passed the pilot test and will be deployed in three more factories: the Audi plants in Brussels and Ingolstadt, and the Volkswagen plant in Emden. However, this also comes with big challenges. The carmaker has discovered that they need a team of experts to train the model again for each factory because the welding configurations are different.
AI can do more than just improve quality control and monitor specific areas. It can also generate valuable data that can help optimize other aspects of manufacturing. Audi is looking into using this data to predict when machines need maintenance, which could save time and money by avoiding breakdowns. The data acquired using AI and the related “predictive” solutions for errors will be used to optimize other production processes as well. This new technology is already being installed in the Audi factory in Brussels and will arrive by the end of the year at the Volkswagen plant in Emden.
As explained by Gerd Walker, Audi Board Member for Production and Logistics, “digitalized assembly lines are a foundation for Audi’s vision for the production of the future: As part of our 360factory production strategy, we will make production at Audi locations worldwide even more efficient. In this regard, the use of artificial intelligence in series production promises huge potential.”
The process had to meet the audit and certification criteria, so the Neckarsulm team joined forces with the German Association for Quality (DGQ) and two Fraunhofer Institutes for Industrial Engineering (IAO) and Manufacturing Engineering and Automation (IPA). This was essential because there are no established standards or certifications for AI applications so far. As the WPS-Analytics project managers Mathias Mayer and Andreas Rieker explained: “This means that, for example, in manufacturing, it is possible to conclusively demonstrate how the result of an AI-based test comes about.”
This project, called “WPS-Analytics, serves as a key use case of the Automotive Initiative 2025 (AI25) launched by Audi. The AI25 aims to create a network of experts for digital transformation and innovation in the factory. It uses digitalization to make production and logistics processes better. Audi and its partners in fintech and IT are testing and using digital solutions in a real factory in Neckarsulm. They are leading the way for the future of smart factories and the efficient making of premium cars.
Tesla is one of the pioneers of AI in the automotive industry with its Autopilot system, which enables semi-autonomous driving on highways and city streets. Tesla also uses AI to optimize its battery performance, charging network, and vehicle software updates. In June, Tesla announced that during the current month it will start production of its supercomputer, Dojo, which is designed to train neural networks for self-driving cars at an unprecedented scale.
BMW is also implementing AI in various aspects of its business, such as design, production, quality control, and customer service processes. BMW also leverages AI to provide personalized and intelligent features for its drivers and passengers, such as voice assistants, gesture control, facial recognition, and driver monitoring. This year, BMW presented its iDrive 8 system, which integrates AI and cloud computing to offer a seamless and intuitive driving experience.
This year, Uber is also applying for a patent for a system that uses AI and machine learning to pair up riders and drivers based on predictions of customers’ preferences. The system would use “context data” from Uber user profiles, such as their past trips or location data, to feed a “machine-learned model” that uses AI to detect patterns in users’ behaviors.
Nvidia is a leader in AI and graphics, and it has been innovating in the car industry with its software and hardware solutions. One of its latest innovations is the Nvidia DRIVE AGX Orin, which is an embedded supercomputing platform that processes data from multiple sensors to enable autonomous driving, with 254 TOPS of performance and support for over 200 trillion operations per second.
With its cutting-edge solutions, Audi is leading the way in the car industry’s transformation by AI. Every aspect of Audi’s factory, from quality control to maintenance to production, is enhanced by AI. AI is also helping Audi establish standards for AI applications in the industry, with the collaboration of its partners and experts. For Audi, AI is not just a technology, but a vision for its future. What will AI do for the car industry in the next decade? What are the benefits and risks for carmakers and drivers? How will AI improve our mobility and lifestyle? These are some of the questions that Audi is addressing with its 360factory production strategy.