Perspectives on the Expertises
Video statement by FIR e.V. on the expertise “Development, use and monetization of an industrial data basis”.“
In the Expertise “Aufbau, Nutzung und Monetarisierung einer industriellen Datenbasis” published in November by the Research Council Industrie 4.0, FIR e.V. at RWTH Aachen and the Industry Maturity Center have conducted research on maintaining competitiveness in the industry. Maximilian Schacht and Tobias Schröer explain the need for the digital transformation of established business models and pose the question of how potentials of industrial data can be better utilised in German companies. You can find an answer to this question, as well as further recommendations for action, in the video statement.
Video statement from FIR e.V.
Maximilian Schacht and Tobias Schröer on the new Expertise of the Research Council Industrie 4.0
By loading the video, you agree to YouTube’s privacy policy.
Learn more
Current positions on the expertise “AI for the implementation of Industry 4.0 in SMEs”
Viewpoints from production research
Prof. Gisela Lanza
wbk KIT and deputy spokesperson science of the Research Council.
“I believe that when you talk about the need for action, you first have to complete digitalisation. AI is then a tool within digitalisation.”
“What does science have to do? Clearly ‘Explainable AI’. And ‘Explainable AI’ not only describes what the computer scientist has to do, but also actually requires translators, so-called production IT specialists.”
“A medium-sized company will definitely only build AI models on the product side,
not on the production side. Because he does not live from production,
he lives from his product.”
Prof. Thomas Bauernhansl
IFF University of Stuttgart,
Fraunhofer IPA and member of the Research Council.
“If you divide AI into perception, cognition and action, then structured data acquisition with cameras or other sensors and evaluating it in terms of ‘OK/not OK’ or anomaly detection and similar approaches is highly advanced.”
“If AI is complex, then SMEs don’t dare to tackle it. That means it actually has to work intuitively, without the need for expert knowledge.”
“I also think that it is very important to develop solutions that can cope with little data. In addition, for the acceptance of an AI solution, especially among SMEs, explainable AI is required. You have to turn the black box into a white box.”
“So what I would find helpful would be young engineers who acquire a lot of knowledge about AI and then become so-called AI technology scouts. I can imagine someone saying: ‘I have the ambition to be at the forefront of AI. I know the technologies.’ Someone who comes to companies – perhaps even government-sponsored – and looks at where there are deficits and potential.”
“An AI trainer would be a really hopeful measure for me. To train a hundred young people in Germany in institutes that say: ‘I am the expert in the field of AI.'”
Prof. Eberhard Abele, PTW of TU Darmstadt – emeritus
“We use AI far too little, especially in the SME sector. When we look at AI applications outside of production and logistics – i.e. in the classic marketing area, trade area, etc. – the Americans and Chinese are light years ahead of us. But if you look at production and logistics, the world looks different. Europe is still ahead in this regard. But now really just the nose. The others have caught up massively. But we are still ahead and this is an opportunity for our industry and especially for SMEs.”
“The biggest starting point or challenge in the area of artificial intelligence and SMEs is making the possibilities and, above all, the benefits understandable, so that someone who does not deal with the topic on a daily basis, who perhaps does not even have the training, can understand the possibilities. Many, especially bosses of medium-sized companies, department managers or production managers who don’t necessarily have studied mechanical engineering, but perhaps have a master’s degree and have grown into their position, do a great job. But they have never had anything to do with modern technologies.”
“The first point that would be necessary to reduce hurdles is a significant improvement in the understanding of what AI can and cannot do. The second is: Either the medium-sized company comes to where they can experience it, or what they can experience comes to the medium-sized company.”
Prof. Wilfried Sihn, IMW at TU Vienna
Prof. Joachim Metternich
PTW of TU Darmstadt
“On the research side, we should work on building AI toolboxes. Using these toolboxes, we can then characterise problems and immediately identify the appropriate AI applications, which we provide preconfigured. And then the user actually only needs to load the data in, make a few explained settings and he gets the output.”
“We need an ‘embedded AI’ –
Digitalisation with AI on board, so to speak.”“You need a personality at the top of a company who looks forward and tries something out. And it takes the willingness to throw what you’ve tried back into the bin. Out of five things we try, one remains really successful.”
I think what helps most are entrepreneurs who have already used AI successfully and who tell other entrepreneurs about it.”