Applied artificial intelligence: Superjoker in production

The EU Commission has presented an ambitious strategy for how Europe can become more independent from the rest of the world, and especially from the market leader, the US, when it comes to artificial intelligence. Industry and research are now called upon to work together to act and achieve these ambitious goals.

What does artificial intelligence offer industry?

According to the Federal Statistical Office, one in five companies in Germany already uses AI technologies, but most frequently generative AI for the creation and analysis of written language and speech recognition or images, for example to support marketing and sales.

As a production technology research institute, we focus on applied artificial intelligence that allows computer-controlled machines, systems, and robots to perform tasks in real production environments. Applied AI offers enormous potential for sustainably improving production processes: lower costs, higher product quality, and increased flexibility and resilience for the entire company.

Intelligent use of artificial intelligence

It is particularly important to us to use artificial intelligence algorithms in the right places and in an efficient manner: we consider each company individually, depending on its size, industry, and unique challenges. To do this, we use proven scientific methods to identify ideal AI use cases and implement suitable processes and algorithms in the production environment.

Building expertise and supporting data science teams within the company is also part of our AI offering. Our long-standing collaboration with software development experts in daily project work provides a solid foundation for this.

Decision support: AI application, machine learning and big data

If you want to find out whether your applications can benefit from the use of artificial intelligence, good information is the key: Which applications already exist? Which algo-rithms can be adapted to your tasks? And how do you train the AI so that it outputs the right results? We know the application areas and solution approaches you can work with.

 

Where is artificial intelligence worthwhile?

Medium-sized companies in particular are often faced with the question of whether the investment is worthwhile. With the "AI Kick-Starter Bundle" we support your decision.

 

Software solution for AI and machine learning in industrial applications

The IQP is a comprehensive software for deploying, monitoring and optimizing industrial AI and Machine Learning applications. Through standardized integration and parallel operation of different ML applications, the IQP increases efficiency and profitability in production. Learn more about the IQP and the use cases that have already been implemented.

 

Optimize processes with AI and machine learning

Artificial intelligence and machine learning help to improve processes. We show you suitable models, systems and architectures for process optimization.

 

 

Trustworthy industrial AI

AI experts have designed a framework and procedure model for the development of trustworthy industrial AI applications that is specifically tailored to the challenges of production technology.

 

Test with free data sets!

For those who want to gain experience with AI and machine learning, we have compiled a list of data and algorithms that you can use for free.

AI implemented: Examples from research and practice

Higher product quality through AI in failure management

Practical experience shows how artificial intelligence (AI) can significantly improve production processes: Together with MAN Truck & Bus and KRONE, we investigated the potential of AI in error management along the value chain as part of the "value chAIn" project and published the methodology we developed in a free paper.

The aim was to provide a methodology that helps companies use AI specifically to ensure product quality. In the project, the team developed specialized AI software that helps companies detect errors at an early stage and efficiently prioritize preventive measures.

This project clearly shows how AI can sustainably improve not only quality but also efficiency in production processes.

"The project has impressively demonstrated that AI has the potential to optimize our production processes in the long term and minimize waste products."


Maximilian Dresemann, Project Manager, Krone Business Center GmbH & Co. KG

"By using AI, we can significantly improve our error management and efficiently prioritize preventive measures to avoid errors at an early stage."

 

Sebastian Beckschulte, Head of Value Stream Planning Cabin Cracow, MAN Truck & Bus SE.

Up to 20 percent fewer unfilled positions thanks to AI-powered workforce planning

In the "reQenrol" project, we demonstrated how artificial intelligence can make personnel planning more flexible and resilient:

The intelligent system considers short-term staff shortages and changes in the production program and efficiently distributes the remaining staff to the available workstations. This not only improves responsiveness to unforeseen events, but also ensures that employees' qualifications are used optimally.

The test run showed that using the system reduced the number of unfilled workstations by up to 20 percent.

The follow-up project "FlexPEP" has already started. The goal here is to support manufacturing companies in making working time models and personnel in production more flexible. The system is being developed and validated in the project using practical use cases.

Intuitive CAM planning with AI

Planning tool paths for machining processes in CAM systems requires in-depth expert knowledge. Many parameters must be determined and checked in order to optimize the path planning step by step. To make this time-consuming task easier in the future, we have teamed up with partners in the "CAMStylus" project to create an AI-supported virtual reality environment for gesture input, in which the tool paths can be intuitively sketched on the workpiece surface.

To ensure that the correct information for the intended tool path can be derived from the motion tracking data, the research team developed an AI application based on neural networks that were trained for this task using specially designed geometric bodies.

Current publications of the Fraunhofer IPT

The Fraunhofer IPT regularly communicates its research results on artificial intelligence in scientific publications. Further publications on this topic can be found in the Fraunhofer-Publica.

Jahr
Year
Titel/Autor:in
Title/Author
Publikationstyp
Publication Type
2025 Optimizing Non-Isothermal Glass Molding Processes: Methodology and Comparative Analysis of Sequential and Non-Sequential Approaches
Mende, Hendrik; Leyendecker, Lars; Upadhyay, Kashyap; Grunert, Dennis; Schmitt, Robert H.
Zeitschriftenaufsatz
Journal Article
2025 Deep learning based automation of mean linear intercept quantification in COPD research
Leyendecker, Lars; Weltin, Anna Louisa; Nienhaus, Florian; Matthey, Michaela; Nießing, Bastian; Wenzel, Daniela; Schmitt, Robert H.
Zeitschriftenaufsatz
Journal Article
2025 Adaptive Control Strategies for Networked Systems: A Reinforcement Learning-Based Approach
Gilerson, André; Bünte, Niklas; Kehl, Pierre E.; Schmitt, Robert H.
Zeitschriftenaufsatz
Journal Article
2025 Automated flank wear segmentation and measurement with deep learning image processing
Holst, Carsten
Dissertation
Doctoral Thesis
2025 A data management system for precision medicine
Jacobs, John J.L.; Beekers, Inés; Verkouter, Inge; Richards, Levi B.; Vegelien, Alexandra; Bloemsma, Lizan D.; Bongaerts, Vera A.M.C.; Cloos, Jacqueline; Erkens, Frederik; Gradowska, Patrycja; Hort, Simon; Hudecek, Michael; Juan, Manel; Maitland-van der Zee, Anke H.; Navarro-Velázquez, Sergio; Ngai, Lok Lam; Rafiq, Qasim A.; Sanges, Carmen; Tettero, Jesse; Os, Hendrikus J.A. van; Vos, Rimke C.; Wit, Yolanda de; Dijk, Steven van
Zeitschriftenaufsatz
Journal Article
2025 Explainable neural network for time series-based condition monitoring in sheet metal shearing
Becker, Marco; Niemietz, Philipp; Bergs, Thomas
Zeitschriftenaufsatz
Journal Article
2025 Enhancing Tool Wear Segmentation with LoRA-SAM and Point Prompts
Li, Zongshuo; Huo, Ding; Meurer, Markus; Panesso Perez, Miguel Antonio; Drossel, Welf-Guntram; Bergs, Thomas
Zeitschriftenaufsatz
Journal Article
2025 A comparison of transformer and CNN-based object detection models for surface defects on Li-Ion Battery Electrodes
Mattern, Alexander; Gerdes, Henrik; Grunert, Dennis; Schmitt, Robert H.
Zeitschriftenaufsatz
Journal Article
2024 Modularity of Manufacturing Systems for Efficient Prototyping
Zontar, Daniel; Rojacher, Cornelia; Hillmer, Nils; Batzel, Christian; Paria, Hamidreza
Zeitschriftenaufsatz
Journal Article
2024 Optimizing Production Lines for Soft and Deformable Products with Agile and Flexible Reconfigurable System
Mazzuto, Giovanni; Ciarapica, Filippo Emanuele; Hellmich, Jan Hendrik; Moya-Ruiz, Laura; Fraile Gil, Francisco
Paper
Diese Liste ist ein Auszug aus der Publikationsplattform Fraunhofer-Publica

This list has been generated from the publication platform Fraunhofer-Publica