Paid Diploma / Master's Thesis

AI-driven self-optimizing DEM simulations
Ref. No. DA192

To dedicated students of electrical/chemical/process/information engineering, mathematics, physics or related disciplines, we offer an opportunity to write a paid Diploma/Master’s Thesis.

OBJECTIVE

Process optimization in Discrete Element Method (DEM) simulations is usually performed manually by adjusting parameters and evaluating the resulting macroscopic behavior. This trial-and-error approach is time-consuming and requires expert knowledge. With the emergence of “Agentic AI” systems, there is the potential to automate and intelligently control such optimization tasks.

Instead of starting with a complex industrial system (e.g., a feed frame), this thesis will focus on a simplified test case, such as a rotating drum or even a two-particle system. The goal is to let an AI-based agent autonomously vary process parameters (e.g., fill height, mass, material properties, drum size) and optimize a defined target quantity, such as the average particle velocity.

The goal of this Master’s thesis is to develop and test a proof-of-concept framework for AI-driven self-optimization in DEM simulations. The outcome should demonstrate whether and how an agentic AI approach can efficiently identify optimal process parameters in simplified DEM scenarios, forming the basis for later application to more complex industrial systems.

WITHIN THE FRAMEWORK OF THIS DIPLOMA / MASTER’S THESIS WE OFFER THE FOLLOWING:

• Extensive participation in a top-level and industrially relevant research project in an international environment
• Supervised training in the task
• Assistance of experienced staff with the implementation of innovative ideas
• Access to highly modern infrastructure on campus of Graz University of Technology
• Assistance with the publication of results

FINANCING

• Compensation on the basis of a service contract

If you are interested in writing your thesis at the interface between university research and industry/ business and to contribute to the optimization of product and process development in the pharmaceutical industry, please apply directly via our website.