"If you want to compete internationally and still be profitable, as a manufacturing company you have to take an interest in the physics that takes place inside the manufacturing processes," says Professor Jesper Hattel from DTU Construct. He elaborates:
"What do materials actually experience when they melt and when they solidify? When internal structures are built up? What is actually happening?"
Jesper Hattel is head of the Section of Manufacturing Engineering, where they work with analysis and optimization of manufacturing processes, including collaboration with manufacturing companies. In December 2024, he was awarded a Dr.techn. based on his many years of work in numerical modelling of manufacturing processes. Numerical modelling is a method of using mathematical models that are programmed into a computer to simulate and analyse complex systems and phenomena, which can be from the world of physics, but also from the world of finance, for example.
To understand how to optimize a company's production, you must know something about the processes needing optimization. To gain this insight, there are two basic approaches: data-driven and physics-driven modelling.
"With the data-driven approach, you either do well-designed experiments that measure in a targeted way or use "observational" data where you log data more randomly by monitoring a production line. The physics-driven approach involves having a more theoretical understanding of what's going on and using that to explain what's happening," says Jesper Hattel.
