The automotive industry is changing rapidly and, with it, the development processes. This is why IAV will be making consistent use of “digital engineering” in future. In the automotion interview, Executive Vice President Stefan Schmidt, who set up IAV’s Digital Lab, describes what is behind it all and the extent to which it is already in place at IAV.
Why is there so much talk about “digital engineering” at the moment?
Stefan Schmidt: Because the automotive industry is being confronted with two megatrends that will also have a major impact on the way it works. We are faced with many different powertrain systems which makes developers’ work extremely complex. In addition, vehicle ownership is no longer so important, particularly for young people, which is why alternatives, like car-sharing or other new mobility concepts, are gaining importance. Everyone in the industry must come to terms with this and develop new ideas beyond the classic business model. And for this, we need digital engineering.
Schmidt: It all starts at the beginning of the development process. In a design thinking phase, attention focuses on the question what exactly consumers need. Which product are they best served by? For this purpose, we need to round up various disciplines at one place and let them work together on finding a solution. In the new mobility world, the end product needn’t necessarily be a privately used car. Often, it’s about getting from A to B, whereby the means of transport can also be a shuttle, a scooter or a shared vehicle. City dwellers’ changed demands and expectations are also playing a part: They want more space for cyclists and pedestrians whereas private cars are tending to become less important. It is aspects like these that we need to consider before we get to work on a specific solution. In our Digital Lab, the experience we have gathered with this approach has been highly positive.
Schmidt: Here, too, we need to work together across boundaries, in this case beyond the boundaries of the domains. We need agile, collaborative work methods. In most cases today we still develop simultaneously and reconcile the results at specific milestones. In future, developers will cooperate through a platform so they can immediately see what a change at one point – such as the chassis – means for the other disciplines, e.g. for the powertrain. For this, we need new IT platforms the developers can dock onto with their domain-specific tools. This way, we can include aspects, such as Internet of Things, software and business models, at an early stage of the development process.
Schmidt: We are already building platforms like these as the basis for interconnecting individual domains. We follow an agile approach, i.e. consecutively add the various domains and learn a lot in this process. Most of the requisite software adapters we develop ourselves. This is one of the reasons why software experts are in so much demand at the moment and why our developers are having to place a greater focus on IT aspects. So, we are not only facing a technical challenge, but also having to move our staff into the age of digital engineering. We are doing this, for instance, by offering a broad range of training opportunities and the capability of working on dedicated project spaces prepared for agile methods.
Which skills must staff have in future?
Schmidt: What IAV really has understood is that the ideas from the Digital Lab provide good models for the entire company. The various divisions and departments are currently looking into what they can quickly adopt. As part of our IAV 2025+ strategy, we have drawn up a road map which will be implemented over the next few years. And we are not alone here: everyone in the industry – OEMs, component suppliers and service providers – are focusing attention on it. Because one thing’s clear: anyone failing to adapt is out.
Collaboration doesn’t stop at company boundaries. Will platform-based collaborative work soon be practiced across the industry?
Schmidt: This is basically what will happen. The extent to which people will work in this way will, for example, depend on how much insight an OEM is prepared to give a supplier into the overall system. But we are well-prepared: our CPU 24/7 subsidiary operates high-security computing centers through which we can connect with our customers for developing vehicle bodies or for working on aspects of autonomous driving. But we are also in a position to dock on to our customers’ platforms.
Mr. Schmidt, thank you for talking to us.
Stefan Schmidt, previously head of IAV’s Digital Lab, has been in charge of IAV’s Project Management Office since the beginning of 2019. This is when Matthias Schultalbers, Executive Vice President of Powertrain Mechatronics, took over responsibility for the Digital Lab.
IAV is a Partner of the Cyber Valley Research Network
Since the beginning of 2017, IAV has been a partner of the Cyber Valley research network, one of Europe’s largest research partnerships from science and business in the field of artificial intelligence. Its aim is to promote dialog between application-oriented industrial research and curiosity-driven primary research.
IAV and the other partners each hold a € 1.26 m interest in Cyber Valley. The money goes into a research fund which, in turn, benefits various research groups at the Max-Planck Institute for Intelligent Systems and the Universities of Stuttgart and Tübingen. Each group, each researcher is free to decide which projects they wish to realize, whether with or without a partner from industry.
Within the research network, IAV as well as scientists from the Cyber Valley “Intelligent Control Systems” group have launched a new research project at the Max-Planck Institute for Intelligent Systems (MPI-IS). Initiated by the head of the research group, Dr. Sebastian Trimpe, and IAV, a new postgraduate project has been set up for which they have been able to attract up-and-coming scientist Alexander von Rohr. Von Rohr and Trimpe have teamed up with other IAV experts to carry out research into self-learning processes in automotive engineering, putting them at the interface between machine learning and control engineering.
Trimpe has worked with IAV before. He is looking forward to further collaboration under the Cyber Valley roof. “In our projects with IAV, we have been able to build a bridge between science and application”, Trimpe says. “Collaboration with IAV works exceptionally well because both sides address fundamental research issues in respect of learning systems and implement their findings on realworld machines.”