When we receive a signal that our body isn’t functioning at 100 percent, we usually work quickly to remedy whatever is causing the problem. The many systems that keep us alive are constantly working together and sending us signals—telling us we need to slow down, rest, put down that second bowl of ice cream—so we can stay healthy and avoid injury.

In the United States, bridges must be inspected, at a minimum, once every two years. Bridges that are in critical condition may be inspected as frequently as every six months but even that may not be often enough as proven by the catastrophic collapse of a fracture-critical bridge in Pittsburgh earlier this year. There are thousands of bridges in poor condition scattered across the United States, each suffering from varying levels of breakdown daily. An inspection every year is great but predicting the speed at which a bridge in poor condition will fail is still an exercise in futility.

“This whole project was developed around existing assets, some of which could be 100 years old. That’s where the focus is needed. Oftentimes, old bridges were overengineered for their time but there have been so many changes over the years—the size, weight and length of vehicles has increased. Traffic volumes have increased. Periodic inspections only give you a snapshot every two years -- like humans, getting a physical every two years. Our bodies change in two years. So do infrastructure assets. The industry needs a real-time, accurate, data-driven information source,” explained Eloque CEO Campbell Rose AM.

A Central Nervous System on a Bridge

Bridge inspections are valuable for assessing the condition and deterioration of a structure but are simply a picture in time. Inspectors and engineers do not receive a continuous stream of data to inform them about how the bridge is behaving during the 364 other days of the year that it is not being inspected. Without feedback, there is no way to evaluate how well a bridge is holding up to the stresses of loading, no way to predict if a catastrophic failure may be on the verge of occurring, and no way to better plan preventative maintenance based on real-time behavioral data.

A solution that makes it possible to understand the day-to-day reactions and behavior of bridge assets would be a godsend for the industry, enabling engineers to make better decisions, predict deterioration patterns and stop failures before they happen, while also providing transportation agencies with the ability to better allocate their resources.

It is Eloque’s mission to develop that solution and the Australian company appears to be making progress.

“Asset management and engineering best practice are ever-moving goalposts,” said Rose. “It is time for these practices to adopt advances in technology. Doing so will assist in addressing the magnitude of the infrastructure problem—this is the core of the Eloque solution.”

Born out of Xerox’s Palo Alto Research Center (PARC), Eloque utilizes fiber-optic wires to feed data into an analytical model that helps interpret the day-to-day behavior and condition of a bridge. Eloque’s first bridge installations have been in Australia, where the company has partnered with VicTrack, the Victorian government’s transportation enterprise, to install remote-sensing and monitoring devices on 12 bridges.

The monitoring devices are little more than hair-thin wires that measure strain, thermal movements, bending due to moment, live load response, vibration and internal corrosion. The sensors transmit data into an advanced analytical model and dashboard that updates in real time and can be viewed at any time by bridge owners. The real-time nature of the data and dashboard allow engineers to react quickly when a bridge has been damaged or to evaluate the impacts of a natural disaster like a hurricane or earthquake. These wires become the central nervous system of the bridge, informing structural engineers of minute changes in the bridge’s behavior and response to loading that could indicate more serious deterioration.

The wires are simple to install on bridges—the process is no more involved than slapping the wires on the underside of the bridge’s superstructure—and can be scaled up quickly to build a comprehensive asset network. The company will first focus on seeing its product installed on older bridges, where the need is currently great, but the fibers could one day become a standard construction item for new bridges.

How Does It Work?

Strain gauges and similar slap-on devices that measure bridge movements have been around for decades but do not offer sophisticated real-time data or actionable insight into the overall behavior of the bridge to the extent that Eloque’s solution can. There are also highly complex devices and systems that can be installed to monitor bridges and transmit data back to a central command center wirelessly. Neither of these options tick all the boxes for widespread deployment across an asset network. The most basic strain gauges can be installed as quickly as Eloque’s wire but lack the same analytical capabilities. Conversely, high-tech devices capable of measuring movements, vibrations or load response are costly and not scalable across hundreds or thousands of bridges.

This is the Eloque solution—a perfect melding of the ease of installing a standard strain gauge with the computational power of complicated measuring devices. It is practical for installation on a large network of smaller rural bridges that are not frequently inspected. These are the bridges that are out of sight, out of mind—unlike iconic structures such as the Golden Gate Bridge or the Sydney Harbor Bridge that are monitored around the clock.

While the solution seems simple, it is anything but. Eloque’s platform works as a result of thousands of hours spent by some of the finest minds in structural engineering and computational methods and artificial intelligence. The Eloque dashboard is a confluence of the rapid advancements in wireless connectivity, data science, AI and structural and finite element modeling. It is a multidisciplinary approach required to solve the most complex challenges in bridge engineering.

Ozan Celik, Eloque’s principal structural analytics engineer, broke down the complex mathematical models that his team has worked tirelessly to develop.

“We are looking at these bridges both at the component level—girders, abutments, decks, girders—as well as global behavior. Since we have so many sensing points on one fiber, we are able to deploy our hybrid system, fiber-optic sensors in a very dense sense right now. That gives us the opportunity to look at each component of the bridge almost like the output of a finite element model where you load your system, you look at your components like a heat map and then see where your stresses and strains are distributed,” Celik explained.

“This is the computer modeling part. What we are doing is looking at it from the actual performance side, because sometimes real life is quite different from the theoretical design. Sometimes the components and the global behavior of the structure can be a little different. There was a gap between the understanding of these two sides, so we are trying to bridge that gap with data taken from actual bridge behavior.”

With enough data, Eloque believes it will be able to predict bridge behavior across typical classes and categories of structure, especially simple-span structures that are the most common. The ultimate goal is to be able to effectively model and predict the behavior of a very high volume of bridges. The pilot projects in Australia have been focused on installing the sensors in a way that most effectively captures bending moment, shear forces and deflections, which will be used to develop key performance indicators that can be used by stakeholders to evaluate their bridges and plan maintenance.

“Those metrics are going to change over time statistically. We can create long-term trends and those long-term trends can be forecasted to feature because your data is coming in continuously. We will have an adaptive system which is continuously creating statistics showing the current condition of your structure in terms of your internal forces. If you are monitoring a crack, how is that crack is propagating?” Celik said.

“We are a young company. These are the first metrics we are looking at. But as we get this data, we are also training models. We are getting into machine learning. We are training our models to make these predictions based on the actual data that’s coming from the specific bridge. Forecasts are being made using statistics at this time but also with informed, trained models. So, in that sense, we are creating a model of that bridge that is continuously updating itself to make predictions for the future or to tell us about the current state.”

The Road Ahead

The need for a solution like the one offered by Eloque is obvious in the bridge engineering and asset management world and the company has a clear runway to continue expanding its growth—first in Australia and then in the United States. The plan is to ramp up deployment in Australia to over 100 bridges by the end of 2022 while launching pilot projects with state Departments of Transportation (DOTs) in the United States.

The technology will first be used on bridges, targeting aging, single-span structures with issues that have the greatest potential to be diagnosed and addressed with remote monitoring. Beyond bridges, there are applications waiting to be built for railroads, highways, mining operations, dams and reservoirs, tunnels and ports—the list goes on. Right now, however, the focus is strictly on bridges. Eloque will also work to develop new monitoring capabilities, such as chloride levels, which is another major concern for concrete structures.

Converting the bridge industry, an industry that is somewhat wary of new technology for inspection, is the next big hurdle for Eloque but Rose believes he has the selling points nailed down. Eloque’s monitoring is not designed to replace routine bridge inspections but rather to supplement them and help them become more refined.

“You may get the data once a week. You don’t need to necessarily monitor every second. You keep your inspection force but you direct it to where you see a bridge that’s got an abnormal strain developing. Rather than go look at the whole bridge, the end-to-end solution can zero in on a specific location of a beam because it seems to be showing more strain than it should be. Our solution gives the inspector some very direct ability to know exactly where to go once they get in the field,” Rose said.

All in all, the Eloque technology is an exciting development for the bridge industry that has the potential to allow transportation agencies to overhaul their inspection processes, better predict maintenance and avoid failures in their older, more remote assets that often do not receive frequent, focused attention. As the world’s infrastructure ages and governments are forced to scrutinize their transportation budgets, Eloque has a much-needed solution to help keep the public safe, reduce wasted spending and target the proper infrastructure maintenance before it becomes too late.