In Depth: How to Make Bridges Safer

Aug. 3, 2007 — Just 13 weeks before the tragic collapse of the Interstate 35 Bridge in Minneapolis, inspectors from the Minnesota Department of Transportation examined the structure. According to the U.S. Department of Transportation, "No imminent dangers were observed."

But the 1967 steel arch deck truss bridge, which rated 4 on the DOT's condition scale (0 = shut down, 9 = perfect), was far from safe. On August 1, the overpass buckled and plunged 64 feet into the Mississippi River, killing at least five people and injuring 60.

For the most part, bridge inspectors do their job with their eyes, scanning literally the nuts and bolts of the structure as well as welded connections for defects.

But this method is growing as antiquated as the bridges themselves, which average between 50 and 60 years old. And even if the best eyes could see every external flaw, it's humanly impossible to detect the extent of corrosion, cracking or fatigue hidden inside.

The Federal Highway Administration conducted a study in the late 1990s looking at the reliability of visual bridge construction, according to Brent Phares, associate director of the Bridge Engineering Center at Iowa State University.

"Those inspections were not very good or reliable. They frequently missed cracks. That points to the potential need for better tech that is easy to use and makes better assessments than we can do visually," Phares said.

In light of that, researchers around the country are developing high-performance concrete, advanced sensors and even robotic surveyors to make the nearly 600,000 bridges in the United States safer and inspections more efficient and accurate.

Not Your Father's Concrete

Since the time of the Romans, engineers have used concrete to build their bridges. In more recent times, they have incorporated steel. But concrete crumbles and steel corrodes.

Researchers at Iowa State University's Bridge Engineering Center are investigating ways to use more durable materials for better bridges. For the first time in the United States, the team constructed a bridge using an ultra-high performance concrete.

The material, developed by the French materials firm LaFarge, contains fine particles of sand, cement and steel fibers in a matrix that is denser and stronger than standard concrete.

"The fail rate of any one element is gradual, as opposed to more brittle material, and that gives you more time to respond," said Phares.

What's more, a beam made from the ultra-high-performance concrete can hold twice the load of beam made from the standard stuff. The scientists used the concrete to build a bridge over Little Soap Creek in southeast Iowa's Wapello County and are currently designing a second one that will go into Buchanan County.

Sensors and Sensibility

Federal law mandates that any bridge spanning more than 20 feet be inspected once every two years. But what happens in the meantime? How extreme heat and cold, traffic accidents, earthquakes, tornadoes, hurricanes and chemicals such as road salt compromise the structure's integrity is often unknown and unmonitored.

"We need faster technologies that allow us to look at a broader scope of the bridge," said Kenneth White, professor and head of the civil engineering department at New Mexico State University.

To address this, White and his team have turned three Department of Transportation bridges in Las Cruces into "smart bridges" by embedding fiber optic sensors into each.

Acting as an architectural nervous system, the fiber optics continually screen the structure for signs of damage or deterioration by shooting out beams of light along the network.

Sensors located along each optical fiber pick up the reflected light and send it to a computer housed in a control box nearby. A program analyzes the signal for abnormalities.

"If some damage occurred to the structure, that abnormality would be picked up by these sensors. If the stress got too high, it would send off an alarm," said White.

For now, the technique is hard-wired with cables running from the fiber optics to the control box. But a wireless version is possible, said White. That would allow engineers working in a central office to receive data from remote locations and could also improve recovery of data should the bridge become damaged and otherwise snap communication lines.

Inspector Robot

Until embedded sensors become a reality, inspectors must still go through the time-consuming, labor-intensive, and sometimes dangerous work of examining bridges.

Current rigging and traffic control measures consume 40 to 50 percent of costs associated with the inspection. Typically, traffic must be shut down, while the inspector, perched in a basket crane underneath the bridge, accomplishes the task.

Howie Choset, associate professor of robotics at Carnegie Mellon University in Pittsburgh, and his team are working on a robot that could cut those costs at least in half.

The team is developing two serpentine-like robots, one named Snoopy and the other Woodstock. The idea is that these kinds of robots could do the dangerous job of inspecting, while people monitor the progress in a safe location nearby.

"The benefit of having a snake is now you don't have to put a person in that cherry picker. That has a profound impact on safety," said Choset, referring to the basket crane.

There are other benefits as well. A digital camera at the inspecting end of the snake would record the examination in real-time, feeding video to a computer display for the engineer to watch. Image processing software would analyze the digital feed looking for flaws.

The images would go into a database and in the event of a collapse like the one on I-35, engineers could retrieve the archives and analyze them for a possible history of structural failure.

Bridge Health, From Birth to Death

By themselves, improved materials, smart sensors and capable robots represent valiant efforts toward safer bridges. But bridges, like any creation, have a beginning and an end, and could potentially be monitored not just during operation, but also from design to demolition.

So think researchers at Carnegie Mellon University's Center for Sensed Critical Infrastructure Research, who are developing sensors and laser technology that would allow engineers and contractors to track materials and their performance over a bridge's lifetime.

The idea is that during construction, thousands of tiny microchips could be embedded in structural components from beams to bolts. The sensors, which track temperature, strain and acceleration, could collect data to ensure the materials are not subjected to conditions that would reduce their quality.

Radio frequency tags could label materials so they are not misplaced, but installed in the precise location according to the designer's plan. Maturity meters embedded in concrete could monitor its strength.

During construction and after the bridge is up and running, a wide-area laser scanner could check the physical locations of the components to determine that they are properly placed and aligned.

"With laser scanning, you can go out, and in an hour you can get a complete collection of points over the structure and do some signal analysis to see if all of the points match up with how it was designed," said professor James Garrett, head of the department of civil and environmental engineering and co-director or the Center for Sensed Critical Infrastructure Research.

In the aftermath of an unfortunate collapse, the same laser could scan the structure, helping to put together a three-dimensional image of the disaster site.

Funding Fiasco

For the I-35 bridge, it's too late. It will take months of analysis, picking through the rubble, watching video, and creating computer models to establish what went wrong. It will cost millions.

U.S. Secretary of Transportation Mary Peters recently announced that $5 million in federal relief would be made available for repairs. And the House of Representatives is authorizing additional funds for repair and to relieve congestion.

The funding woes go much deeper. According to the American Society of Civil Engineers, it will cost $9.4 billion a year for 20 years to eliminate bridge deficiencies across the country.

"It's one of those seriously underfunded endeavors that deserves more attention," said Choset.

In the meantime, accidents like the one on I-35 and recently the collapse of 1-70 in 2005, which injured 70 people in Pennsylvannia, may continue to happen.

"I think that what happened on I-70 and this incident on I-35 are going to call attention to the need to do more for our bridges," said Garrett.

The hope is that technology can fill that need.

The Department of Transportation's fact sheet on the I-35 collapse.

Design News investigates the disaster.

The Center for Sensed Critical Infrastructure Research

Smart bridges

The Bridge Engineering Center at Iowa State