Erdzan Hodzic

Erdzan Hodzic is a senior researcher at RISE Materials and Production and received his PhD in Computational Mechanics from Lund Institute of Technology (LTH, Lund University) in 2016. The research mainly revolves around the use of classical numerical methods and machine learning in a variety of manufacturing processes. In addition, he is engaged in exploring and implementing machine learning and artificial intelligence in the manufacturing sector.

On the one hand, the computational research focuses on using detailed computations and simulations for the study of the processes related to, melting and solidification of different materials, characterization of instabilities and other undesirable physical phenomena during manufacturing, as well as optimization of other processes related to fluid dynamics. Through these large-scale calculations and optimization of both the process and the product, the quality of final products is improved, efficiency is increased, and the risk of defects and rejects is reduced.

The second part of the research deals with the application of machine learning (ML), including physics-based/informed neural networks (PINNs), Neural Operators (NO), symbolic regression, and various AI/ML-based methods within the area of physics-based machine learning (Phys-ML). These methods have shown great potential to automate, improve/optimize and in some cases even replace the classical computational methods for solving Partial Differential Equations (PDE:s), saving time and opening new opportunities for solving complex multi-physics phenomena in industrial applications.

The common thread are the fundamental challenges in physics, focusing on technology- and application development for computations and simulations, machine learning and statistics within the field of materials and production, conducting research with both academia and industry across several different fields and projects.