Argonne National Laboratory has introduced SMART-NDI, an artificial intelligence-driven inspection framework designed to enhance non-destructive inspection (NDI) processes and improve quality assurance across advanced manufacturing industries.
The technology, formally presented as “SMART-NDI: Scalable Manufacturing Assessment and Real-Time Testing for Non-Destructive Inspection via Artificial Intelligence,” utilizes a multi-task, self-supervised deep learning framework to identify manufacturing defects more efficiently than traditional inspection methods.
A key feature of SMART-NDI is its ability to detect anomalies using only non-defective data during training. Unlike conventional AI-based inspection systems that require extensive libraries of labeled defect examples, the framework learns normal material and component characteristics and subsequently identifies deviations that may indicate flaws or defects.
According to Argonne National Laboratory, this approach addresses one of the major challenges associated with AI-enabled inspection systems, where building comprehensive defect databases can be both time-consuming and resource-intensive. By eliminating the need for curated defect libraries, SMART-NDI offers manufacturers a faster path to deploying automated inspection capabilities.
The framework has been developed to support a broad range of non-destructive inspection applications and can be adapted to different materials, manufacturing processes, and inspection techniques through the integration of relevant datasets. Its modular architecture enables users to customize the system for specific industrial requirements.
Argonne researchers reported that SMART-NDI has been validated in industrial environments, including aerospace inspection applications, where it demonstrated high levels of defect detection accuracy while reducing inspection times. The technology is designed to streamline inspection workflows, reduce reliance on manual review, and improve production efficiency.
The laboratory noted that traditional quality assurance procedures often create bottlenecks in manufacturing due to labor-intensive inspection processes. By automating anomaly detection and accelerating inspection cycles, SMART-NDI has the potential to reduce operational costs, lower energy consumption, and support increased production throughput.
Beyond aerospace, the system is being positioned for use across sectors including automotive manufacturing, microelectronics, infrastructure, and critical materials production, where inspection reliability and compliance with strict safety standards remain essential.
Further research on the technology was presented at the International Conference on Machine Learning and Applications through the paper titled “DOC-DICAM: Domain Aware One Class Defect Identification in Composite Aerostructure Material,” highlighting the framework’s application in composite material inspection.
As manufacturers continue to seek scalable inspection solutions capable of supporting growing production demands, SMART-NDI represents a significant development in the application of artificial intelligence to non-destructive inspection, offering a flexible approach to defect detection without dependence on extensive defect datasets.
Reference: https://quantumzeitgeist.com/argonne-laboratory-ai-powered-inspection-cuts/
