Abstract
Numerical simulation has long served as a foundational pillar of innovation in manufacturing, enabling the systematic design, evaluation, and optimization of complex engineered products. Yet despite its maturity, the increasing complexity of industrial systems continues to challenge traditional high fidelity modeling workflows, raising concerns about their long term sustainability and scalability. In parallel, the broad and often transformative narrative surrounding artificial intelligence has generated both enthusiasm and anxiety within the simulation community. This seminar aims to critically examine the impact of scientific machine learning on the landscape of simulation-based engineering. Beyond presenting recent methodological advances, we will address the persistent challenges associated with integrating data-driven tools into established physics-based pipelines. Emphasis will be placed on understanding not only what these techniques can accelerate or automate, but also the extent to which their adoption can generate tangible, measurable impact in industrial product development processes.
