The Impact of AI on the Automotive Sector

From robots on the factory floor to autonomous cars on the road, AI is transforming the automotive sector.

The new technology has plenty of room to expand, increasing efficiency, productivity, and safety throughout the process of automotive manufacturing. In the next five years, AI is expected to completely revolutionize the industry.

“I think much of AI in manufacturing has been limited to predictive maintenance and in the next five years, there will be a much wider set of use-cases of AI in manufacturing as its value is better understood,” Frans Cronje, CEO of DataProphet, told Generis Group.

Ahead of the American Automotive Summit, we spoke with Frans to discuss the impact of AI on the automotive industry, what steps manufacturers need to take to leverage the technology, and what the future of AI looks like.

How is AI impacting the automotive industry?
AI is having a large impact on the automotive sector, with great attention on self-driving cars. However, our focus is on applying AI to the production of vehicles. Here, we see AI as part of Industry 4.0 initiatives, driving up efficiencies in manufacturing plants by improving overall equipment effectiveness (OEE), reducing defects, and improving automation on the line. These benefits are achieved using AI solutions that support predictive maintenance, active optimization of control parameters, and computer vision.

How can AI help improve key plant metrics, reduce downtime, rework, and scrap?
It is important to understand the different use cases and value adds that AI projects can provide to the line. Automating visual inspection helps reduce human error in the process and improve traceability, however, it does not reduce the chance of a defect occurring, only the chance of one being shipped. Predictive maintenance can help with overall efficiencies in the plant, however, we see that the greatest immediate value add in production is in the optimization of control parameters. By actively optimizing control parameters with AI, production can expect a reduction in defects to occur, reducing the cost of non-quality.

Control parameters are often configured and only reviewed after a defect occurs, and even then the review is often done with traditional analysis techniques that can not work on the full data available. Therefore, correcting the control environment is done very reactively which can actually increase the variance of quality on the line. AI can prescriptively assign optimal control parameters to reduce the chance of the defect ever occurring and help reduce quality variance on the line. This result is achieved because it is not limited in the volume of data that AI can work with and AI can actively work on the current data available to predict how the future quality will be impacted, and then produce proactive actions to steer production to the optimal production.

What are the steps manufacturers need to take to leverage AI?
AI is a value add to data. In order to leverage AI, it needs to be enabled by data. This means that a manufacturer needs to have a good data environment or a route to a good data environment. What we emphasize is that most of the data collection hardware is already installed. Capital equipment that was installed in the last 20 years will have a good set of sensors on them. However, the collection of that data is important, especially from a holistic point of view.

We work with many clients to improve their data environment to reach a state where they can leverage AI. We put a lot of emphasis on AI for Industry 3.0 rather than Industry 4.0, which speaks to working with the current set of sensors on the line and creating value from the data they produce. It is often not necessary to add more data streams before the existing ones are ordered and value is created from them.

One further important point we like emphasizing is that real-time data is often too late. Real-time data is only valuable if you can respond in real-time. In practice, a good AI solution should be able to act in advance of real-time. I think this is well-understood in predictive maintenance but often not in the optimization of control parameters.

In reality, prescriptive actions need to take into account how many changes can be made to the line and how fast they can be made, especially when these recommendations are provided to the control engineers.

How can DataProphet optimize production in the automotive industry?
We provide two solutions. The first, DataProphet PRESCRIBE, is AI for optimizing control parameters. Here we have helped customers reduce casting defects significantly, in some cases reducing shipped defects to 0% for periods of up to three months and reducing robotic weld defects by 50 – 70% in body shops.

Our second solution, DataProphet INSPECT, is AI for automated visual inspection. Here, the solution helps customers automate the visual inspection of the surface of items.

We support these two solutions through the digitization services we offer, where we work with our customers to help them achieve the data environment they need to realize the value from their data.

Where do you see AI in the manufacturing sector in the next five years?
I think much of AI in manufacturing has been limited to predictive maintenance and in the next five years, there will be a much wider set of use-cases of AI in manufacturing as its value is better understood.

Your session at the American Automotive Summit will focus on how AI-enabled process parameter optimization can assist in reducing defects and reworks. What is one key takeaway you want to leave attendees with?

I have two key takeaways I hope to leave attendees with. The first is that current data streams are more than sufficient to start extracting value from AI. What we see across most industry strategies is to aggregate data, and this should really be done with the value adds upon the data in mind. The second is that real-time is often too late. In the process control environment, it is important that information is received well before action is taken upon it.

To learn more, visit the DataProphet website or the American Automotive Summit agenda.