The collapse of Baltimore’s Francis Scott Key Bridge in March 2024 was the consequence of a long-standing problem: the fragility of aging infrastructure. As reconstruction gets underway on an estimated four-year timeline, the disaster reflects the urgent need for better bridge inspection nationwide.

The U.S. is home to more than 600,000 bridges, many of which have supported heavy traffic for decades. While steel is renowned for its strength, constant exposure to the elements leads to corrosion, gradually weakening structural integrity and increasing public safety risk.

More than 200,000 U.S. bridges require major repairs, and 46,000 are classified as structurally deficient, according to a 2024 ARTBA report. The repair backlog alone is estimated at $125 billion, highlighting just how critical efficient inspection is. Yet traditional methods are increasingly falling short.

A smarter approach to bridge inspections

For decades, bridge inspections have relied on visual checks and ultrasound measurement—methods that are slow, expensive, and prone to human error. Certain structural components are also difficult and dangerous to access, leading to lengthy inspections which often require lane closures that disrupt traffic.

To bridge this gap, researchers at University of Massachusetts-Amherst and the Dresden University of Technology have developed a new inspection method that blends 3D scanning with AI-driven predictive analytics. Their approach has already been funded and tested in collaboration with local authorities in Massachusetts.

Steel bridge girders showing signs of deterioration (reflected in the captured 3D scan data). Credit: Artec3D

Digitizing infrastructure with 3D scanning

At the heart of this breakthrough is the Artec Leo, a wireless, high-speed 3D scanner capable of capturing steel beams with unparalleled accuracy. Unlike traditional tools that measure a single point every few minutes, Artec Leo enables researchers to collect hundreds of thousands of data points in just five minutes.

The scanner’s AI-powered software, Artec Studio, then transforms raw data into high-resolution, photorealistic 3D models that capture every detail, including reflective metal surfaces.

These models act as digital blueprints that enable engineers to critically analyze structural integrity without the limitations of conventional methods, onsite or anywhere else. With this technology, inspectors can rapidly detect beam deviations, corrosion hotspots, and material loss, all without scaffolding, lane closures, or lengthy downtime.

Data-driven insights for smarter decision-making

Scanning is just the beginning. Once collected, the 3D data undergo analysis using finite element analysis software. This simulates how corroded beams respond under different loads, and it pinpoints failure risks with a level of accuracy that conventional inspections simply can’t achieve.

To further accelerate efficiency, researchers have integrated machine learning. Their AI model, trained on thousands of corrosion scenarios, can now analyze scan data and predict beam capacity with remarkable precision. This automation removes much of the guesswork from structural assessments, providing engineers with immediate, data-driven insights to determine a bridge’s true level of safety.

A complete girder 3D scan, including small holes and fine surface details. Credit: Artec 3D

This technology has already proven itself in the field. During a Massachusetts bridge inspection, researchers leveraged this AI-enhanced workflow to navigate accessibility challenges, manage environmental vibrations, and generate actionable insights in record time.

For infrastructure authorities, this streamlined approach could help agencies like the Federal Highway Administration prioritize maintenance, extend the lifespan of steel bridges, and ultimately make smarter, cost-effective investments in public safety.

Scaling innovation for the road ahead

For widespread adoption of this technology, integrating standardized data collection methods into existing bridge management systems will be key. Some states are already taking initial steps by acquiring Artec Leo scanners for real-world testing.

As momentum builds, AI-powered workflows have the potential to revolutionize infrastructure management, shifting agencies from reactive repairs to proactive, data-driven maintenance. This could lead to a future where AI-driven load rating and corrosion tracking are fully embedded into national bridge databases, creating dynamic digital records of structural health.

Such a shift would enable agencies to continuously monitor bridges in real time, anticipate maintenance needs, and make targeted investments that extend infrastructure lifespan while optimizing repair budgets.

By making AI-driven monitoring a cornerstone of national infrastructure strategies, we can ensure the safety, efficiency, and longevity of our bridges. This could be transformational for how we manage and sustain the vital transportation networks connecting us all, and ultimately create a safer, sturdier infrastructure for generations to come.

To learn more about this pioneering research, read the full paper here.

Published by Artec 3D, a global leader in 3D scanning technology.