10/14/2025

Public lecture series: ‘Data-based AI and machine learning for engineering sciences’

Zwei junge Männer an einem Tisch mit einigen Instrumenten und einem großen Bildschirm
© Oliver DietzeSteffen Klein (left) and Christopher Schnur from Professor Andreas Schütze's team have collaborated on the development of a maintenance system that combines artificial intelligence with sensors that collect condition data from machines: The system independently assigns signal patterns to damage, wear or fault conditions, thus making the condition of a plant permanently visible and learning new conditions from the data during operation.

Artificial intelligence is not only a topic in computer science, but also indispensable in engineering. An upcoming lecture series in the winter semester 2025/26 will highlight how engineers use AI, especially machine learning and data-based methods in general for industry, using basic lectures and examples from research.

The public lecture series is organized by the Systems Engineering department at Saarland University and will take place every Thursday from 23 October, 4:00 to 5:30 p.m., on the university campus (Building C4 3, Lecture Hall 0.02). The lectures are open to everyone. 

The following text has been machine translated from the German with no human editing.

Data is the basis of the modern information society – this applies to social networks as well as to industrial production, for example. Whether welding technology, crash simulation or robotics: sensors that collect a wide range of data, integrated computers for data processing, and algorithms and models for drawing the right conclusions are always involved. Engineering sciences not only use physical models to describe the behaviour of systems, but increasingly also so-called data-based approaches. These observe system behaviour on the basis of measurement results and, after a learning phase, can make predictions and optimise processes.

The current lecture series covers the basics of data, the necessary data quality and processing, machine learning and deep learning, as well as general methods for optimisation and data processing that run directly on smart sensors without cloud data centres. In addition, specific data-based applications are presented as examples, including the description of shape memory alloys, welding processes and improved passive vehicle safety, as well as soft robots based on dielectric elastomers and their use in non-destructive testing (NDT).
The series will conclude with two lectures with demonstrations at the Center for Mechatronics and Automation Technology (ZeMA, 29 January 2026) and at Fraunhofer IZFP (5 February 2026), which will focus specifically on the fields of robotics and the use of large language models (LLM) such as Chat-GPT for industrial applications.

The lecture series is aimed at students and research assistants at the university as well as users in industry who want to learn about the basics and key concepts of AI/ML or who are planning specific applications themselves and are looking for inspiration or cooperation partners. Teachers, trainers and the general public are also invited to attend the lecture series.

Further information:
All dates and a link to participate online via Teams can be found at:
https://www.uni-saarland.de/fachrichtung/systems-engineering/perspektiven-der-ingenieurwissenschaften.html

Fragen beantwortet:
Prof. Dr. Andreas Schütze
Lehrstuhl für Messtechnik
E-Mail: schuetze(at)lmt.uni-saarland.de
Tel.: 0681 302-4663

A press photo can be found on this news website. You may use it free of charge in connection with this press release and reporting on Saarland University, provided you credit the photographer.

Zwei junge Männer an einem Tisch mit einigen Instrumenten und einem großen Bildschirm
© Oliver DietzeSteffen Klein (left) and Christopher Schnur from Professor Andreas Schütze's team have collaborated on the development of a maintenance system that combines artificial intelligence with sensors that collect condition data from machines: The system independently assigns signal patterns to damage, wear or fault conditions, thus making the condition of a plant permanently visible and learning new conditions from the data during operation.