The Department of Energy’s Oak Ridge National Laboratory (ORNL) has published a new set of additive manufacturing datasets aimed at transforming quality assurance in 3D printing. The release is expected to provide manufacturers and researchers with tools to evaluate part performance without relying solely on traditional destructive or costly non-destructive testing (NDT) methods.
The publicly available dataset is hosted on an online platform designed to give what ORNL describes as “a complete story” around additively manufactured components. By capturing data during the printing process rather than relying exclusively on post-production analysis, the approach enables predictive insights into part quality.
Over the past decade, ORNL’s Manufacturing Demonstration Facility has collected extensive data on 3D printed components, pairing advanced manufacturing research with performance testing. This information is now being used to train machine learning models that can predict material performance based on in-process measurements. According to ORNL, the trained algorithm demonstrated a 61% reduction in errors when predicting a part’s ultimate tensile strength compared with conventional approaches.
Vincent Paquit, head of ORNL’s Secure and Digital Manufacturing section, explained the significance:
“We are providing trustworthy datasets for industry to use toward certification of products. This is a data management platform structured to tell a complete story around an additively manufactured component. The goal is to use in-process measurements to predict the performance of the printed part.”
Traditionally, additive manufacturing quality has depended on destructive tensile testing or advanced non-destructive evaluation (NDE) techniques such as X-ray computed tomography. While effective, these methods are often expensive, time-consuming, and less suited for larger parts. ORNL’s dataset—spanning 230 gigabytes—covers the design, printing, and testing of five sets of laser powder bed parts with varied geometries. It includes machine health sensor data, laser scan paths, 30,000 powder bed images, and 6,300 tensile strength tests.
Paquit described the dataset as a “key enabler to additive manufacturing at industry scale,” allowing manufacturers to “capture the link between intent, manufacturing and outcomes.”
This marks the fourth and most extensive dataset ORNL has made available to the public. Developed under the Advanced Materials and Manufacturing Technology Program, funded by the DOE’s Office of Nuclear Energy, the effort aims to accelerate advanced manufacturing adoption for reliable, cost-effective nuclear energy applications.
By combining real-time process monitoring with machine learning and NDT approaches, ORNL’s dataset could become a cornerstone in certifying 3D printed parts for critical industries, reducing reliance on destructive testing and enabling predictive quality assurance at scale.
