Kostenlos abonnieren

Werden Sie regelmäßig per E-Mail über neue Ausgaben der campuls informiert. Sie können Ihr kostenloses Abo jederzeit einfach online über den Abmeldelink im Newsletter kündigen.

Weitere Infos zu Datenschutz & Widerrufsrecht finden Sie hier.

Industry project at iisys: More competitiveness through forward-looking installation management

Since November 2021, the Institute for Information Systems (iisys) at Hof University of Applied Sciences has been researching artificial intelligence models for industrial companies. These are intended to provide companies with precise information about the condition of their machines. Predictive maintenance”, i.e. anticipatory maintenance, can thus prevent production waste. The Cyber-Physical Systems research group is supported in the project by AMITRONICS Angewandte Mikromechatronik GmbH from Munich in the form of high-performance sensors on a real-time measurement system combined with intelligent evaluation methods, as well as by Scherdel GmbH in Marktredwitz, which is making its plants and planning data available for research purposes. The research project is scheduled to run for a total of 3 years. It is funded by the Bavarian Joint Research Program (BayVFP) with a sum of over 900,000 euros.

Mobile UMTS development environment for rapid measurement data acquisition and development of anomaly detection in continuous operation. 1.Intelligent vibroacoustic measurement system, 2. powerful industrial LTE router, 3. acoustic emission sensors installed on an electric motor, 4. development environment; source: AMITRONICS Angewandte Mikromechatronik GmbH;

Especially in the automotive industry, dynamic global competition, rising energy and raw material prices, and a volatile automotive market are currently causing cost pressures that can no longer be met by conventional means. Savings potential therefore lies, among other things, in avoiding rejects from production parts, i.e., eliminating defects entirely during production or at least detecting them at an early stage.

AI technologies for predictive maintenance and quality monitoring

Based on structure-borne sound, e.g. humming or vibrating, statements can be made about the condition of machines. This requires an intelligent, real-time measurement system in the area of sound emission and coordinated company processes. This is supplied by AMITRONICS from Munich and installed on the machines of Scherdel GmbH.

Our high-performance sensors can identify quality-influencing anomalies on manufacturing equipment or on manufactured products through acoustic emission signals.”

Andreas Hofer, head of research at AMITRONICS Angewandte Mikromechatronik GmbH

As soon as irregularities occur in the data evaluation, Prof. Dr. Valentin Plenk’s research group comes into play:

Based on the measured structure-borne sound, we can see what state the machine is in. We can also make a statement as to whether the fault is perhaps even due to the processing material or the tools.”

Prof. Dr.-Ing. Valentin Plenk, Vice President for Research and Development at Hof University of Applied Sciences

This is made possible by a real innovation: Scherdel GmbH in Marktredwitz not only makes its plants available for research, but also its complete planning data – i.e. information on which production parts were manufactured on the machines at any given time. The research group evaluates sound and production data together and can thus tell exactly what kind of defect is present and where.

As a company from Upper Franconia, we are very proud that the Hof University of Applied Sciences is researching new methods in our manufacturing facilities to optimize production processes.”

Dr. Johann Haertl, Concept Development and Strategy in the Department of Development and Technology Company Scherdel GmbH

And further: “If the research group can prove that this process is really promising, we can well imagine transferring this technology to all our operating equipment in the future.”

Increasing the competitiveness of companies

The Hof University of Applied Sciences’ AI models are thus intended to provide transparency as to what condition the respective machine or fleet is in. Employees already receive information about potential problems during their shift. Production errors or risks can thus be detected at an early stage and the causes eliminated in real time. The end result is an increase in yield, predictability and thus the competitiveness of the company. Classic types of quality assurance can thus be completely replaced.


Franziska Brömel

Weitere Themen