The magazine "ProgSim" (Progressive Simulation) is the German-language online magazine for engineering simulation focusing on Artificial Intelligence (AI), Machine Learning (ML), Simulation Process and Data Management (SPDM), Computer Aided Engineering (CAE) / numerical simulation methods and related fields.

AI / ML
Artificial Intelligence and Machine Learning are becoming increasingly integrated into engineering simulation. Current developments include surrogate and reduced-order models, physics-informed machine learning, geometric deep learning and agentic AI. Key questions remain around data requirements, model validity, uncertainty, traceability, validation and the level of autonomy that should be given to AI-based engineering workflows.

SPDM
Simulation Process and Data Management provides an increasingly important foundation for AI-enabled engineering. Structured, traceable and reusable simulation data can support the development and validation of AI/ML models, while SPDM environments can provide controlled workflows in which models, processes and engineering agents operate with appropriate traceability and governance.

CAE
Traditional physics-based simulation remains an essential part of engineering decision-making while becoming increasingly connected with data-driven methods. Interoperability between CAE tools, SPDM and PLM systems, optimisation environments, data platforms and AI agents is therefore becoming an important aspect of modern simulation workflows.

Contents

Probabilistic AI for Engineering: Evaluation of a Single Model Family Across Tabular, Mesh-Based and Particle-Based Benchmarks
Kevin Cremanns (PI Probaligence)

Datenfreie Physics-Informed Neural Networks zur Modellierung industrieller Strömungen
Amin Zargaran, Uwe Janoske (Bergische Universität Wuppertal)

KI-gestütztes Simulationsdatenmanagement für industrielle Entwicklungsumgebungen
Maciej Klebowski (CAIQ Polska), Alexander Mahl (CAIQ)

Von Daten zu Entscheidungen: Eine Graph-KI-Architektur für die dialoggestützte Crash-Analyse
Sebastian Fink (GNS Systems)

Datenmanagement als Basis für die Verknüpfung von Simulationen und physischen Tests
Sebastian Schweigert-Recksiek (:em engineering methods), Roland Kummeth, Alwin Perras, Steffen Hartmann (Schaeffler Technologies)

Premium sponsors of this issue:

PI Probaligence

Simr