"Life‑cycle–driven evaluation is the ultimate gatekeeper for determining whether additive manufacturing delivers real economic and environmental value at scale"
Key takeaway: life‑cycle–driven evaluation is the ultimate gatekeeper for determining whether additive manufacturing delivers real economic and environmental value at scale..png?width=1170&height=777&name=Repam%20Hybrid%20Event%20Resource%20Efficiency%20in%20Additive%20Manufacturing%20%20(2).png)
During the event key presentation topics were:
- Socio-Economic Value and Health & Safety in Circular Additive Manufacturing – Luc Plancke
- Resource efficiency in additive manufacturing processes
- Circular material flows and sustainable feedstock use
- Monitoring, sensors, and data collection for process understanding
- Recycling routes for scrap, chips, powders, and wires
- Artificial intelligence and modelling for process control
- Industrial relevance of circular additive manufacturing
"Metal Additive Manufacturing Technologies: Trends, Research, Economy & Future Perspectives"
Oguzhan Yilmaz, Professor of Additive Manufacturing (Gazi University), presented "Metal Additive Manufacturing Technologies: Trends, Research, Economy & Future Perspectives"
He covered:
- Definition of additive manufacturing: building objects from 3D model data, usually layer‑by‑layer, vs. subtractive methods.
- Metal AM process categories: melting‑based and solid‑state metal AM.
- Global 3D printing market overview.
- REPAM research topic distribution: sustainability, IoT in AM, life cycle assessment, AI in AM, metal powder reuse and recycling.
- Economic insights for metal AM: focus on steels (e.g., 316L, low‑alloy), recycling, sustainability, lifecycle, and IoT.
- Strategic ranking of REPAM objectives: life cycle assessment (highest economic impact), followed by AI in AM, sustainability, material efficiency, and IoT/digital AM—noting LCA as the ultimate gatekeeper for economic and environmental value at scale.
- Future trends: micro‑ and nanoscale AM and applications across electronics (MEMS), pharmaceuticals, medical, aerospace, and defence.
"From monitoring to insight: sensors and data for AM processes"
Samira Gruber (Fraunhofer IWS) presented "From monitoring to insight: sensors and data for AM processes" covering:
- Why sensors in AM: enable real‑time monitoring and closed‑loop control, hyperspectral imaging (HSI) for powder qualification.
- HSI scope in REPAM: detection of qualitative and quantitative powder changes.
- HSI hardware: VEpioneer CORE as an at‑line, compact analysis system and the Powder Container.
- HSI data: raw data handling and analysis.
- HSI applications: hyperspectral imaging of metal powders to detect aging (virgin → aged), relevance for reuse and production, next steps to calibrate an Aging Index and set acceptance thresholds (go/no‑go).
- Recoater sensors: monitor/detect print errors, assess powder characteristics during recoating, enable early adjustment or cancellation of failed jobs; recommended basic and additional sensor data to acquire.
Next steps: output‑integrity prediction, analysis and preprocessing, feature development/extraction, and feature selection.
"Artificial Intelligence in Additive Manufacturing"
Dr. Sachin Salunkhe, Professor of Department of Mechanical Engineering, presented on "Artificial Intelligence in Additive Manufacturing."
He covered the following topics:
- AI fundamentals and applications for AM, including Industry 4.0 integration and real‑time detection and prediction of part defects.
- AI decision‑making versus perception (right: perception-based; left: rational-based) and introduced NLP as an AI method for communicating with intelligent systems.
"Digital process intelligence: modelling and AI for process control and optimization in REPAM"
A four‑member team Mohammad Hoseini-Athar, Victor Lamelas, Maddi Etxegarai and Chris Hulme presented "Digital process intelligence: modelling and AI for process control and optimization in REPAM"
They reviewed models for centrifugal and gas‑atomization processes, noting:
- The fine tail is the most predictable target, with stable behavior and consistent model performance.
- The coarse tail is hardest to predict due to very low variability, making modeling and evaluation sensitive to noise.
- D50 shows intermediate results; models capture trends but risk overfitting from weak signal‑to‑noise and approximation error.
- Adding modelling features for surface tension and viscosity improves predictions for fine and coarse tails, while using process + modelling features (without chemical composition) benefits D50 predictions but not the other targets.
Through events like the REPAM hybrid meeting, Enperas will continue participating in the project to support knowledge exchange, stakeholder engagement, and awareness of sustainable additive manufacturing in Europe and beyond.
