Answers are grounded in this paper and cite their sources.
The project DatProForge (Data-driven Process Modelling of Drop Forging Processes to Increase Productivity Using Adaptive Tool Design Methodology), conducted within the first funding period of SPP 2422 has established the foundations for an end-to-end digital process chain combining experimentally acquired process data, automated finite element (FE) simulations, radar-based condition monitoring, and data-driven and explainable artificial intelligence (XAI). The resulting methodology provides the basis for developing an intelligent assistant system for process design and optimisation. The intelligent assistant system employs machine learning (ML) and XAI methods to identify defect risks, determine their key influencing factors in a transparent manner, and generate recommendations for process optimisation and adaptive tool design. Using a hot bulk forging process on a servo screw forging press as a representative use case, the paper demonstrates how research results on sensor systems, radar signal processing, stochastic simulation, underfill prediction, vent-channel optimisation, and data-pipeline modelling can be integrated into a unified assistance system. Particular attention is given to the joint analysis of simulation results and experimentally acquired process data to identify relevant process and machine-related effects and to account for associated uncertainties. The resulting simulation-to-reality transfer approach enables the FE models to be successively refined by incorporating these effects, thereby progressively improving their agreement with the real forging process. Combined with interpretable data-driven models, this approach provides a foundation for robust, data-driven decision-making in process optimisation and adaptive tool design.
No takes yet. Share an insight, caveat, or question.
Härtel et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: