Requirements for numeric models as sources of synthetic data for predicting real-world data sets in progressive deep drawing processes

Abstract In the field of forming technology, synthetic data generated by finite element (FE) simulations is increasingly being used to train machine learning (ML) models for part quality prediction. However, the predictive accuracy on real-world process data is often limited by the so-called “reality gap” between simulated and measured signals. This study investigates how simulation model complexity influences the suitability of synthetic data for training ML models that generalise to real progressive deep-drawing processes. Three representative simulation configurations of increasing complexity (L1–L3) for a symmetric part are implemented and evaluated against experimental data collected under production conditions using sensor-integrated tools. The analysis covers multiple ML tasks, including classification of pre-process connector cut geometries, detection of process disturbances, separation of subtle geometry variants, assessment of feature transfer robustness, and saliency-based interpretation of signal regions. The results show that simple simulations (L1) enable robust classification of failures, such as material damage within synthetic domains. However, their transferability to real data is limited. While computationally more expensive, higher complexity levels (L2 and L3) better capture the effects of pre-processing and deformation history, improving domain alignment and supporting physically meaningful model interpretation. Saliency analysis reveals that models trained on synthetic data emphasise different signal regions than models trained on real data. This underscores the importance of task-relevant signal fidelity. The findings provide quantitative guidance for selecting adequate levels of simulation complexity in comparable progressive deep drawing applications.