A research team led by Jeong Min Park from the Korea Institute of Materials Science in collaboration with Jaemin Wang and Dierk Raabe of the Max Planck Institute for Iron Research, has developed an artificial intelligence based model capable of predicting the likelihood and characteristics of internal defects during metal additive manufacturing process design.
The breakthrough is expected to significantly enhance the reliability of metal 3D-printed components and accelerate their adoption for large-scale industrial production.
Metal additive manufacturing has emerged as a next-generation manufacturing technology capable of producing complex and high-value components. However, its wider industrial application has been constrained by microscopic internal defects that occur during production, often leading to component failure and performance degradation. Traditional quality evaluation methods typically rely on simple indicators such as porosity, but the actual impact on mechanical performance depends on factors such as the shape, size, location, and distribution of defects.
To address these challenges, the research team developed an Explainable Artificial Intelligence model that systematically analyzes and predicts the relationships among process conditions, defect morphology, and mechanical performance in metal additive manufacturing. This approach allows manufacturers to predict potential defects and their performance impact at the process design stage, enabling a new framework for defect-aware process design and quality management.
A key feature of the AI model is its ability to analyze internal defects generated during the laser powder bed fusion process by evaluating morphological characteristics such as shape, pore size, non-circularity, and spatial distribution. By analyzing microstructural images, the model automatically correlates these defect characteristics with mechanical properties, providing a quantitative explanation of how defects influence performance.
Unlike conventional “black-box” AI systems whose decision-making processes are not transparent, this explainable AI model identifies why certain process conditions increase defect formation and degrade mechanical performance. The research team trained the model using extensive datasets that included process parameters, powder characteristics, defect imaging data, and mechanical performance results from multiple materials such as steel, aluminum alloys, and titanium alloys.
Through this integrated framework, the model can stepwise predict how process variables influence defect formation and how defect morphology subsequently affects mechanical performance.
The technology is expected to significantly improve the quality reliability of metal additive manufacturing and accelerate its adoption in industries that require highly reliable components, including aerospace, defense, and mobility. By reducing defect rates as well as material waste and rework costs, the model can also enhance overall industrial production efficiency.
Dr. Jeong Min Park, the lead inventor from KIMS, said:
“This research goes beyond simply reducing defects in metal 3D-printed components; it establishes a scientific framework that explains how specific types of defects directly influence performance. We expect this work to contribute to the broader industrial adoption of metal additive manufacturing, particularly in high-performance sectors such as aerospace, space, and defense.”
The research was supported by the KIMS Fundamental Research Program, the Materials and Components Technology Development Program funded by South Korea’s Ministry of Trade, Industry and Energy, and the Energy Efficiency Innovation Technology Development Program. The findings were published online on January 1, 2026, in Acta Materialia, one of the leading global journals in metallurgy.
The research team plans to expand the technology into a digital twin-based quality management system that can be deployed in real-world industrial manufacturing environments.