“100 per cent logistics”: that is the motto of the Fraunhofer Institute for Material Flow and Logistics (IML) in Dortmund, Germany. Together with Deutsche Telekom, the logistics experts founded “Telekom Open IoT Labs”. “Logistics is the driving force of the 4th Industrial Revolution and the Internet of Things”, declares Prof. Dr. Dr. h. c. Michael ten Hompel, managing director of the institute.
Professor, what are the goals of the joint venture with Deutsche Telekom?
The arrangement represents a new form of cooperation between industry and the Fraunhofer Institute. Together with other companies, we operate several such labs. With Telekom Open IoT labs, we want to make the Internet of Things become reality. To do this, we need full-coverage, ultra-low-power wireless communication, i.e. a very energy-saving solution in the right frequency range. I am convinced that the wireless technology Narrowband IoT
(NB-IoT) is the appropriate standard for this. This “Low Power Wide Area” (LPWA) technology combines low power requirements with excellent building penetration and wide area coverage.
Prof. Dr. Dr. h. c. Michael ten Hompel
In the Open IoT labs, though, is it not just strictly about research?
Absolutely. Our mission at Fraunhofer is to promote industrial development. Thus, we conduct less basic research; our main focus is the industrial implementation of innovation. In the Open IoT labs we will develop turnkey solutions – as quickly as possible. We have set very specific objectives for ourselves to achieve this.
You believe that NB-IoT will prevail as a logistics standard?
In addition to the inherent advantages of low energy consumption and wireless connectivity even through walls, narrowband IoT is also an element of the LTE standard. And, it is fit for the future, because it will be set forth in the next generation mobile communications standard 5G. The logistics sector needs such solutions, because we transport things and are always on the go. Battery-powered applications, for which we need to constantly change the batteries, are not practical.
However, no large amounts of data can be transferred via NB-IoT.
This is not a primary need of the logistics sector. We do not stream any videos. With us its about item information, identification numbers, inventories. The bandwidth is therefore not decisive. In logistics its about kilobits and not the gigabit range.
But a wireless standard alone still doesn’t constitute an IoT solution.
I have been concerned with the Internet of Things for many years and I can state: we now have the technology available that we need for IoT. This applies not only to the network, but also to the sensors and computing power, e.g. in the form of highly-parallel GPUs – the storage technology that enables us to actually impart all things with intelligence. For example, load carriers, receptacles, boxes, crates, containers, swap bodies or vehicles.
You want to affix foils to storage boxes. What is the purpose of this?
They are highly flexible displays that were developed by us, with which we transmit information to logistics employees who are standing in front of a transport item. What's in the box? What should they take?
Additionally, an automated replenishment capability can also be implemented. The crate then reports once a specified threshold has been gone under and reorders independently.
What is the challenge for logistics and how can IoT help to overcome the challenge?
It is primarily a matter of transparency in the supply chain. Transparency is the first step in the 4th Industrial Revolution. A logistics specialist must understand what happens in his or her supply chain. First of all, this means receiving data, then interpreting it by means of a cloud-based application, and then planning more accurately and quickly on the basis of this data. Ultimately, the goal will be that things start to autonomously (to a certain extent) control themselves.
Capturing and transmitting the data is technically no longer a critical issue. What happens next to the data?
The complexity in logistics often stems from highly scalable applications. We’re not talking about 1,000 containers – we’re talking about 100,000 containers or perhaps even millions, whose data we must route through the networks. This is not at all a trivial matter. Because, in fact we need to refine the raw data; for that, we need to set up algorithms that run in the cloud
. Based on the analysis results, we can then reschedule operations, such as the routing within logistic networks.
Are artificial intelligence methods behind this?
This cannot yet be characterized as artificial intelligence, but nonetheless algorithms from general AI research and data science are being applied. For example, in the field of logistics this could mean: if certain conditions are present, then you had better not take the A1 highway but instead you should drive via the A2. Analytical methods such as these are related to concepts such as Reinforcement Learning, Big Data
and Predictive Analytics. In the next step, they form the basis for applying AI algorithms, such as Deep Neural Networks.
There are already logistics companies who track their trailers and containers, as well as their status. With IoT-Labs, what does Telekom want to add to this?
Today's solutions utilize the existing networks, i.e. GSM or LTE. Comparatively, they consume a lot of energy – which is accordingly apparent in today's tracking systems. They have a relatively large battery on board. However, we are talking about operating periods of five years or more. The issue of energy consumption plays an enormous role in this context. If we now equip the load carriers with Low Power Wide Area (LPWA) technology, we take a decisive step ahead and can then ponder all sorts of other developments.
The logistics sector is open to innovations of this kind?
In the past, if I told someone at a bar that I am a logistics specialist, then they would ask me where I had parked my truck. In the meantime this has changed quite a bit. Logistics is increasingly on the way to being perceived as a technology industry. This has a lot to do with development pertaining to Industry 4.0, as the examples very often come from the logistics sector. Whether it is the autonomous truck on the road, the mobile robot going through the warehouse, or the smart box. On the other hand, however, it is the case that expenditures for research and development in the logistics sector still are not very high. This is in clear contrast to the claim that the logistics – in every sense of the word – should be moving things. We therefore also repositioned ourselves with the Fraunhofer IML and are happy that we, together with the Deutsche Telekom, are staking out a leading position in the technology development.
Traditional industrial companies have already completely automated their production chain. But supposedly there is still a disconnect to logistics?
With regard to Industry 4.0, we speak of the horizontal networking, including things that are on the move. In terms of data, we speak of information logistics. Logistics stands for the horizontal networking – which we also superbly manage. Production companies, on the other hand, dominate vertical networking. As such, automobile production in Germany is already extremely well organized. And now we must unite these two networks – the horizontal and vertical – which is exactly what we are focusing on with our joint lab. There we expect to benefit greatly in terms of flexibility and adaptability, and ultimately to achieve high efficiency gains.
Are there any figures on which time or financial benefits this will bring?
No, no one on Earth has them. Why does no one have any numbers yet? Because the 4th Industrial Revolution is only now taking place. Only now are we in the midst of constructing the true Internet of Things. Many think that Industry 4.0 has long arrived. But that's not the case. The basic technologies are mature. But the real horizontal and vertical networking has yet to take place. Of course, everyone assumes that we will achieve high efficiency gains while at the same time gaining flexibility. This is a recognized hypothesis. But we are only in the start-up phase.
You are also founders or co-founders of the competence center for small and medium-sized enterprises SME 4.0. Practically relevant solutions and projects are also being carried out there. Is all of this structurally interdependent?
The competence center should provide a blueprint for targeted action. The aim is to take Industry 4.0 to small and medium-sized enterprises and to advise them on how to integrate it. It can only be in the interest of both partners – Telekom and Fraunhofer – to approach these companies as early as possible and work together with them to develop business models.
Prof. Dr. Dr. h. c. Michael ten Hompel
The electrical engineering graduate is holder of the chair of the Materials Handling and Inventory Management faculty at the University of Dortmund and managing director of the Fraunhofer Institute for Material Flow and Logistics (IML), as well as being department head at the Fraunhofer Institute for Software and Systems Engineering (ISST). Ten Hompel is a member of acatech (the German National Academy of Science and Engineering) and the Logistics Hall of Fame, and has an honorary doctorate from the University of Miskolc in Hungary.