13u30 – 14u15 Reliable and Energy Efficient Connectivity for the Industrial IoT

door Jeroen Famaey, imec & University of Antwerp

Bio: Jeroen Famaey received his M.Sc. and Ph.D. degrees in computer science engineering from Ghent University, Belgium, in 2007 and 2012. He is an Assistant Professor associated with imec and the University of Antwerp, Belgium. He has co-authored more than 100 articles published in international peer reviewed journals and conferences, as well as 10 patent applications. He leads a team of researchers focusing on performance analysis and optimization of emerging wireless network technologies for IoT and 5G applications.

Inhoud sessieA wide range of wireless connectivity standards and technologies are (becoming) available for connecting industrial IoT devices to the Internet (e.g., BLE, NB-IoT, LoRa, Wi-Fi HaLow). Although they all aim to provide reliable and low-power connectivity to IoT devices, they vary greatly in terms of coverage range, throughput, energy consumption, latency, etc. In this talk, we will provide the audience with a quantitative and qualitative comparison of current and emerging technologies for industrial IoT connectivity, allowing them to make an informed decision on which technology to use for their specific use cases. Moreover, we will briefly discuss future trends in the area of wireless connectivity for industrial IoT applications.

Model Predictive Control and/or Digital Twins

door Peter Hellinckx, imec & University of Antwerp

Bio: Peter Hellinckx obtained his Master in Computer Science and his Ph.D. in Science both from the University of Antwerp in 2002 and 2008 respectively. is the head of the department of Electronics-ICT and the post graduate in Internet of Things. He is part of the imec research group IDLab at the university of Antwerp. He currently supervises 12 PhD’s, 4 post-docs and a development team in the field of distributed artificial intelligence. . He is currently teaching third year bachelor courses advanced programming techniques, Artificial Intelligence and distributed systems and the master courses IoT Distributed Embedded Software and Computer Graphics. He is co-founder of the spin-offs Hysopt, Hi10 and Digitrans. His research focuses on Distributed Artificial Intelligence for Iot and Cyber Physical Systems with as main application domains: autonomous driving/shipping, logistics, mobility, Industry 4.0 and smart cities. In this field is a reviewer in many scientific project evaluation commissions, both on a national and an international level . He is a guest editor of multiple journals acting on these topics. In the last 5 years 25 of his industrial oriented research projects, national as well as international were funded.

Inhoud sessie: Digital Twin is for sure a buzzword of this time. And as for many buzzwords there is a lot of misinterpretation of its actual meaning. It can and probably should be used for sales purposes but I my talk I want to give you my vision on what it really is and how we can use this in maintenance and control. We will start with a quick history of where it comes from, give an inside in how IoT, AI and HPC changed this world and then focus on its actual added value At the end I will zoom in on the different application domains within design, building, control and maintenance.

14u30 – 15u15 A Smart Maintenance Living Lab for tackling challenges towards industry 4.0

door Steven Devos


  • How to take steps towards maintenance 4.0?
  • Is low-cost condition monitoring feasible?
  • How to deal with the lack of datasets for validating condition monitoring solutions?
  • How to set up a cloud environment for maintenance 4.0? What to process on the edge and what in the cloud?
  • Which processing to use to transform a large amount of data into actionable information?
  • How to visualise the data from the cloud to the user?


Maintenance is often done periodically, where sometimes components are replaced, even if they still function well. Because sensors are becoming cheaper, it’s becoming more interesting to plan the maintenance based on the health of the machine (condition-based maintenance). We will show a case on which we have demonstrated that the assessment of the health of a bearing is feasible using low-cost hardware components.

However, an important bottleneck to roll out solutions on a large scale, is that componanies lack the confidence that these solutions will have a high return on investment. They lack the measurement data on which they can validate these solutions. Rich datasets are needed on which faults are occuring, in a wide range of operating conditions, arnd where it is known what happened.

In order to make a first step to bridge this gap, we have built the Smart Maintenance lab, where we have 7 setups for accelerated life tests on bearings. We focus on bearings because these are critical machine components which are used in many machines. Although the bearing itself does not cost much, the standstill and the escalation of damage due to a failing bearing can be huge.

Using accelerometers, the health indicators on all setups are calculated locally (to reduce the amount of data) with our algorithms. These health indicators are sent to the cloud and visualised on the dashboard. For the whole chain, we are using open technologies so that they can be optimised and integrated for a particular industrial application.

See also the following movies:


  • Start capturing data early or bootstrap on available datasets in orer to validate condition monitoring solutions
  • Health monitoring with with low-cost hardware components is demonstrated for a bearing case
  • The Smart Maintenance lab illustrates on small-scale how steps towards maintenance 4.0 can be taken

In de kijker


Maintenance en Pumps & Valves onder één dak, dat is vuurwerk!

onze partners

Driving technology for leading brands
Scroll naar top