At the centre of the AI system there is a graphical interface, which highlights abnormalities in spot welds, marked by the AI algorithm. Image credit: Audi |
Recently, Audi announced that it has conducted a pilot
project at its Neckarsulm site in Germany. It is using artificial intelligence
to control the quality of spot welds in high-volume production.
The aim of the project is for the algorithm to evaluate
close to 100% of the set welding points during body manufacturing in the future.
Cars are built by robots. A modern car factory is a huge technological facility, where people work in close proximity to robots, but a lot of work is still carried out by specialists.
For example, quality control of spot welds that hold car
bodies together. Audi is looking into automating at least part of that process
and improving the quality control by implementing an Artificial Intelligence
(AI) solution.
Thousands and thousands of spot welds go into a modern car
body. For example, even though Audi A6 is not the largest car out there
(although not a small one either) there are around d 5,300 spot welds joining
its body parts together. They are applied by robots, but then they need to be
inspected for quality. A bad spot weld can cause some issues with the body
integrity, safety, comfort or noise, depending on the location of that spot
weld and the severity of the defect. Therefore, Audi uses ultrasound to
manually monitor the quality of resistance spot welding. As checking all of
those welds is virtually impossible, they use the method of random.
Audi has been conducting a pilot AI project to test
abilities of AI to control the quality
of spot welds in high-volume production. In particular, spot welds in the
production of A6 and A7 car body manufacturing at the Neckarsulm site have been
tested using AI. In the future Audi hopes to
evaluate close to 100 % of the set welding points during the body
manufacturing process. In the future, AI-powered quality control systems will
be fully integrated into the welding processes. It will be controlled fully
automatically and will be optimised continuously. Of course, the result should
be an even lower likelihood of defects in mass-produced car bodies.
Michael Haeffner, Head of Delivery Management Digitalization
for Production and Logistics at AUDI AG, said: “As a pilot plant for digital
production and logistics at the Volkswagen Group, our goal is to test and
develop digital solutions for vehicle manufacturing right through to their use
in series production. With the use of AI, we are testing an important key
technology here that will make Audi and the location fit for the future.” In
the future AI application in Volkswagen Group plants will be hugely expanded.
For example, AI will be looking for quality defects such as small cracks in the
car body.
Artificial intelligence is not new in industrial
manufacturing. AI is often used to optimise logistics, search for improvements
in efficiency, controlling the flow or materials, etc. AI is particularly
useful in quality control, because an algorithm can analyse a huge amount of data
very quickly and produce comprehensible results.
0 comments:
Post a Comment