This article is based on an official press release from Argonne National Laboratory.

Argonne National Laboratory and Spirit AeroSystems Unveil AI Tool for Aerospace Inspections

A new collaboration between government research facilities and private industry aims to resolve one of the most persistent bottlenecks in modern Commercial-Aircraft manufacturing: the inspection of composite materials. On January 7, 2026, Argonne National Laboratory (ANL) announced the development of an artificial intelligence tool designed to accelerate the analysis of ultrasonic scans, a move they report will significantly reduce production time and energy consumption.

The project, led by ANL in partnership with Spirit AeroSystems, Northern Illinois University, and Texas Research Institute Austin, utilizes advanced machine learning to assist human inspectors. According to the laboratory’s announcement, the tool reduces human inspection time by 7% and cuts facility-level energy usage by approximately 3% per aircraft.

Addressing the Composite Challenge

Modern aerospace engineering relies heavily on composite materials due to their superior strength-to-weight ratios. However, verifying the structural integrity of these materials is a data-intensive process. Manufacturers typically use ultrasonic non-destructive testing (NDT) to scan components, generating massive datasets that human experts must manually review to identify defects.

Argonne National Laboratory describes this manual review process as time-consuming and mentally fatiguing. To address this, the research team utilized the Argonne Leadership Computing Facility (ALCF), a U.S. Department of Energy Office of Science user facility, to develop a solution based on Convolutional Neural Networks (CNNs).

According to the technical details released by the laboratory, the AI model does not replace human inspectors. Instead, it functions as an intelligent assistant that rapidly processes scan data to highlight “regions of interest.”

“It rapidly processes scan data and highlights specific areas that contain potential defects or anomalies. This allows human experts to focus their attention solely on ‘regions of interest’ rather than reviewing empty or flawless data.”

— Argonne National Laboratory Announcement

Operational and Environmental Benefits

The integration of this technology offers measurable gains in both efficiency and Sustainability. Spirit AeroSystems, a major aerostructures manufacturer, provided the proprietary dataset of ultrasonic scans used to train the model. The resulting tool has demonstrated the ability to shorten the overall production flow time.

Efficiency Gains

By automating the initial screening of ultrasonic data, the tool reduces the time human inspectors spend on each component by 7%. In high-volume manufacturing environments, this reduction allows for increased throughput and helps alleviate production backlogs.

Energy Reduction

Perhaps most notably, the efficiency gains translate directly into energy savings. The announcement states that the tool lowers energy use by roughly 3% per aircraft. This reduction is achieved at the facility level; shorter inspection times mean that heavy machinery, HVAC systems, and lighting operate for fewer hours per unit produced.

AirPro News Analysis

The deployment of this AI tool highlights a critical shift in the aerospace sector’s approach to “Industry 4.0.” While much of the past decade’s innovation focused on physical Automation, such as robotic drilling or fastening, the current frontier is digital automation.

We observe that the bottleneck in composite manufacturing has shifted from layup (placing the material) to verification (proving the material is safe). As aircraft designs become increasingly complex, the volume of NDT data is outpacing human capacity to review it. The “human-in-the-loop” approach taken by Argonne and Spirit AeroSystems is significant because it mitigates the regulatory hurdles associated with fully autonomous inspection. By keeping the human inspector as the final authority, manufacturers can likely integrate these tools faster than if they sought to replace the human entirely.

Furthermore, the “open-framework” nature of the underlying techniques, mentioned in the release as being available for academic research, suggests that this methodology could soon expand beyond aerospace into wind energy and automotive sectors, where composite usage is also rising.

Partnership Details

The success of this initiative relied on a multi-sector collaboration. While Argonne provided the supercomputing power and machine learning expertise, Spirit AeroSystems supplied the domain knowledge and real-world data necessary to train the AI effectively. Northern Illinois University and Texas Research Institute Austin contributed to validating the technology’s robustness and reliability.

Rajkumar Kettimuthu, a Senior Scientist and Group Leader at Argonne, emphasized the collaborative nature of the work in the official release, noting the combination of industrial constraints and high-performance computing.

Sources

• Argonne National Laboratory