Bielefeld/Paderborn/Lemgo. Under unusual conditions, the first milestone meeting for the project took place on March 25, 2020. Due to the current Corona situation, the previous research results and future steps were discussed online in order to bring machine learning closer to small and medium-sized enterprises in an industrial context.
In the ITS.ML project, inIT is currently working together with four other research partners, CORLab of Bielefeld University, SiCP of Paderborn University and ISyM and CfADS of Bielefeld University of Applied Sciences, on the sustainable availability of machine learning for Industry 4.0. The vision: establishing machine learning and artificial intelligence as a service for small and medium-sized enterprises. To this end, the partners are combining cutting-edge research with concrete application examples from industry, making innovations available and thus providing an entry point for technology transfer to companies.
But where can machine learning methods be used everywhere? Can artificial intelligence be used to predict unplanned downtime in my company due to lack of maintenance? What solutions have already been tested in practice and what value is created by predictive maintenance?
The digitization of plants and their equipment with corresponding sensor technology includes the potential to intelligently control condition monitoring and maintenance of machines through digital sensor values of the plant. Since exact modeling is usually not possible here, approaches from the field of machine learning often provide an optimal solution.
"To be clear, machine learning or artificial intelligence is not magic, but a combination of algorithms, each combination with its own strengths and weaknesses," explains Anton Pfeifer, a research associate supervising the ITS.ML project. "It is becoming increasingly important that we are able to sort through the hype and reality around the topic of artificial intelligence and understand what AI can and cannot currently do." InIT Institute Director Prof. Volker Lohweg adds, "We need AI that is transparent and, especially in the industrial environment, certified with traceability of what has been learned and how."
You can learn how machine learning concepts can be used in a meaningful way in our workshop on predictive maintenance - predicting when a system will fail in terms of preventive maintenance, among other things. The date for the workshop will be adjusted according to demand based on the current situation. More information will follow.
More information about the project at:
Questions will be answered by:
Anton Pfeifer
inIT - Institute Industrial IT
Tel: +49 (0) 5261 / 702-5203
Email: anton.pfeifer@th-owl.de
Dr. Ulrike Kuhl
Cognitive Interaction Technology (CITEC)
Tel: +49 (0) 521 / 106-12246
Email: info@its-ml.de