Is AI a critical tool or a buzzword? Several pilot cases are conducted in the automotive industry

Is AI a critical tool or a buzzword? Several pilot cases are conducted in the automotive industry

ECG — 2025-06-06

News from ECG

Christophe Deville, Toyota Motor Europe, and Anike Murrenhoff, Fraunhofer-Institute for Material Flow and Logistics in Germany, discuss opportunities and challenges.

The Fraunhofer Institute conducts applied research in partnership with industry across all areas of logisticsAnike Murrenhoff was one of the speakers at the ECG Spring Congress in Cascais, Portugal in May.

"AI is enabling many efficiencies in today's processes, and development is going on rapidly," says Anike Murrenhoff. She explains that AI not only automates workflows but also integrates and operates various tools across systems.

"We still rely heavily on manual work, not just on the shop floor, but also in control tasks on a technical, strategical level where we control our logistics systems like supply chains."

How can FVL companies use AI in their daily business?

"One area in which AI is currently being used in industrial settings is perception. Perceptive AI can help get information from the physical world into the virtual world. So thus helping to digitalize processes."

"For example, when you're using your smartphone and taking a picture of a document, the information is understood and taken into your IT system. Similarly, cameras on your shop floor or in your yards can capture data such as the location of handling units—and feed it into databases for further AI applications."

Anike Murrenhoff states that there are some hesitations among small and medium-sized businesses about using AI, as they aren't sure where to start.

What are the challenges?

"A major challenge is acceptance. It's essential to take the employees on the journey with you, get them involved early on in the process and have them talk about the challenges they are facing in their projects or their processes in general. And let them know how AI can help solve these problems."

Another challenge, according to Anike Murrenhoff, is the know-how and expertise in companies. Providers with strong experience in AI can bridge this gap and offer tailored solutions.

Toyota Motor Europe is piloting AI, and Jean Christophe Deville, Vice President of Supply Chain, sees enormous potential in its application.

"The main challenge is not the AI itself or even the technology providers: it's us, our workforce and our ability to work in a more digital world. We are not digitally comfortable with AI yet. The challenge for us is how we educate and upskill our people.”

”The second challenge is that AI reveals its full potential when there is a true collaboration among the supply chain's partners, from the logistics partners to our factories and dealers. We need to better connec and align."

Toyota Motor Europe is conducting a few pilot cases. Jean-Christophe Deville, what are the key takeaways or successes so far?

"The true value lies in how AI contributes to our mission: delivering vehicles ‘On time, In time, Every time’ for every customer. If AI helps us achieve that, it’s welcome. But if it’s just a buzzword with no real business impact, then its worth remains unproven."

"We have a few pilot cases where we have some high expectations. For example, in vehicle inspections, the system is learning from existing data to improve accuracy. But these are still in early stages—they’re not yet validated successes."

All new technologies come with risks, and AI is no exception.

Anike Murrenhoff:

"For sure, in the beginning, for many AI applications you will have to train an AI system for your specific use case, and you will have to have a human check if the results that the AI system is providing are the correct ones. And then, after a while, you will get a feeling about which cases can be solved by AI themselves and which ones a human should look at."

So, what is the best practice to start when you are new to AI applications in logistics?

"I would say to start by thinking of a case that is currently a challenge in your system, then go ahead and find a small proof of concept you can do. Don't start immediaely with a big AI project; find something small you can solve using AI and then eventually build it up to a bigger project", says Anike Murrenhoff.