Prof. Wolfgang Wahlster, CEA of DFKI, one of the founding partners of the RICAIP Centre, is regarded as one of the fathers of “Industry 4.0”. In an interview with the magazine TecniCall of the Czech Technical University in Prague, he talks about how the term was created and what it stands for.
Photo: Jim Rakete, DFKI
The interview was prepared in cooperation with the team of the RICAIP Centre, the English version provided and published by DFKI.
When and with what goal did you first define the term “Industry 4.0”?
We have already conceived a future project of importance for Germany for the use of cyber-physical systems and the Internet of Things on behalf of the Federal Government in the Research Union in 2010.We have established it within the area of cyber-physical production systems, since every second job in Germany depends directly or indirectly on the production of physical products. In December 2010, at a meeting at the German Academy of Science and Engineering (acatech) in Berlin where I met the colleagues Henning Kagermann and Wolf-Dieter Lukas, I suggested that the future project be named “Industry 4.0”. The designation “4.0” heralds the fourth industrial revolution and at the same time gives an indication of the important future role of software through the type of version designation known in IT.On April 1, 2011, our first article on Industry 4.0 appeared in VDI-Nachrichten, Germany’s leading business weekly on industry and engineering, with the title “Industry 4.0: With the Internet of Things on the Way to the Fourth Industrial Revolution”. On June 15, 2011, I distributed my graphic on the four stages of industrial revolutions for the first time on the Internet, which was then translated and adapted by thousands of colleagues worldwide. The intuitively understandable graphics undoubtedly contributed to the rapid spread of the term “Industry 4.0”. The global triumphal procession of the term also led to the fact that in many countries the German spelling “Industrie 4.0” is still used correctly today.
Did you have preliminary work for Industry 4.0 on which to base the concept?
Yes, in specifying the characteristics for Industry 4.0, we were able to build on concrete research results from DFKI, which have been developed since 2005 in our world’s first Smart Factory in Kaiserslautern. On the basis of the demonstrator operated by colleague Detlef Zühlke, I have been leading the SemProm project on semantic product memories since 2008, which assigned a digital twin to each factory module and each product produced. This twin stores the function and the entire life history of a physical object in a machine-understandable representation and actively controls the production process itself, so that mass customization up to batch size 1 is economically possible. I was able to successfully complete the BMBF-funded project in 2011 and publish a book at Springer academic publishing, which is still the essential basis for Industry 4.0 today. At that time, we were already able to produce customized liquid soaps in different mixtures on a fully functional Industry 4.0 system in a service-oriented architecture at DFKI. So we knew what we were talking about when we created the term “Industry 4.0”.
Did you already believe at that time that the concept would be a global success and when will the fourth industrial revolution be completed?
I was already convinced at that time that Industry 4.0 would bring the breakthrough for the Internet of Things in the Factory, especially with the advances in my special field of Artificial Intelligence. The idea ignited like a rocket and is still very much alive. After all, industrial revolutions cannot be completely realized in a decade, but usually take at least 30 years. Today, only a maximum of 10% of all factories in the world have converted to Industry 4.0 and it will certainly take until 2030 until more than half of all factories worldwide are working according to the principles of Industry 4.0. Today, there are still factories in some countries of the world in which work according to the Industry 2.0 standard without any digitalization at all.
You see industrial artificial intelligence as a catalyst for the next generation of smart factories. What are the main reasons for this?
Artificial intelligence is the avant-garde of digitalization. The first wave of digitization of industrial production has already been completed in innovative companies: Data is now only stored, transmitted and processed digitally. They are machine-readable, but not yet machine understandable. The second wave of digitization is based on artificial intelligence technologies. AI software can now for the first time interpret machine data in terms of content. Machines understand the people in the factory and their intentions. Machines can semantically analyse the linguistic descriptions of the mechatronic functioning of machines and thus autonomously deal with new modules. The automatic knowledge acquisition of Industry 4.0 can be achieved by AI in three ways: through deep learning on machine data, through language understanding of technical documents and through speech dialog with engineers and skilled workers.
What other trends will determine the next phase of Industry 4.0?
5G and then 6G cellular networks will enable wireless communication under real-time conditions with guaranteed latency, making remote control of robots and manufacturing equipment and analysis of sensor data over long distances affordable and universal. At DFKI we have successfully implemented 5G in several test environments. This also makes possible so-called edge clouds. We can use them to merge sensor data on site in the factory and analyse them with machine learning methods, e.g. to enable a quality control step-by-step in every production step. Edge devices are no longer cumbersomely wired, but wirelessly networked via 5G technology.A future industry 4.0 factory must run in a Many-Cloud architecture and, for reasons of technical sovereignty, must also be able to switch to federated edge clouds whenever this appears necessary for security or political reasons.To enable interoperability of the multitude of networked modules in Industry 4.0, international standards and norms for the new AI components are now urgently needed. Under my leadership, we have now completed the world’s first AI standardization roadmap together with DIN and DKE on behalf of the German government, which has been published in November. Norms and standards ensure global market access and can also represent a decisive competitive advantage for small and medium-sized enterprises.
How has the global competition for Industry 4.0 markets developed?
Germany is still the world’s most important factory supplier when it comes to the highest quality standards and innovation. Europe as a whole still has a 2-3-year lead in Industry 4.0 over China, Japan and the USA in this area. The good cooperation in Europe with the German pioneers of Industry 4.0 also contributes to this. An excellent example is the Czech Republic, where Prof. Vladimir Marik and his team focused on Industry 4.0 very early and very consistently. With his internationally recognized expertise in the field of holonic multi-agent systems, he has also made a fundamental contribution to the realization of the new distributed and service-oriented production architectures. With CIIRC, CTU is today one of the most important centres of excellence for Industry 4.0 with an excellent application environment in Czech factories. The cooperation between DFKI and CIIRC as well as with INRIA in France, TNO in the Netherlands, DTx CoLabs in Portugal and FBK-ITC in Trento has ensured that Europe remains the leading provider and driver of innovation in Industry 4.0. Our joint EU project RICAIP is an international beacon of Industry 4.0.
Has the Corona crisis also produced relevant findings for Industry 4.0?
Yes, it became very clear how quickly global supply chains can break in a pandemic.This makes the aspect of supply chain resilience much more important in the future. AI-based tools for real-time monitoring of the stability of supply chains and intelligent replanning in logistics can make a significant contribution to this. At DFKI, we were able to implement monitoring and rescheduling at Chinese suppliers for several large manufacturers in Germany, which greatly limited the damage caused by corona-related production downtimes.
What is your advice to young students of engineering sciences?
Studying at CTU gives you the best chances for the future if you specialize in industrial AI. There will continue to be a great demand in the industrial and academic sectors in this sector over the next 10-20 years. With the renowned CTU chairs in Robotics, Machine Learning, Mechatronics, Sensor Interpretation and Multi Agent Systems, Internet of Things and Software Technologies, CTU offers the best conditions for an interdisciplinary study of world-class industrial AI.
The interview has been published in TecniCall (Autumn 2020) first.The questions have been asked by Eva Doležalová and Tilman Becker, RICAIP – Research and Innovation Centre on Advanced Industrial Production.