Latest / Internet of Things with Fexingo: Connected Devices, Sensors, and Industrial IoT / How IoT Sensors Are Making Bridges Smarter
Transcript
- Lucas: You know that sinking feeling when you drive over a bridge and it's clearly not in great shape? Maybe you see exposed rebar or a crumbling edge? Luna: Oh, I avoid those bridges whenever I can. But sometimes you don't have a choice. Lucas: Right. And the scary part is, even a well-maintained bridge can fail without warning. That's exactly what happened with the Fern Hollow Bridge in Pittsburgh back in 2022. It collapsed early in the morning — luckily no one died, but several people were injured and a bus ended up precariously hanging over the ravine. Luna: That was a huge story. And I remember the investigation found that corrosion had severely weakened the steel, but no one caught it in time. Lucas: Exactly. And that's the problem we're diving into today: how IoT sensors can give us a continuous, real-time picture of a bridge's health — not just a visual inspection every two years or whatever the schedule is. Luna: So instead of waiting for cracks to become visible, the sensors detect the tiny changes that happen long before failure. Lucas: Exactly. The technology is called structural health monitoring, and it's been around for a while in labs, but now with cheaper wireless sensors and better battery life, it's actually becoming practical for real bridges. The Fern Hollow collapse became a catalyst for several pilot projects. Luna: What kind of sensors are we talking about specifically? Lucas: There are three main types. Accelerometers measure vibrations — so if a bridge starts to oscillate differently, that's a red flag. Strain gauges measure how much a beam is bending or stretching under load. And tilt sensors detect if a pier or abutment is shifting out of alignment. All of them can be wireless and send data to a cloud dashboard. Luna: And what does that data look like in practice? I imagine a lot of noise. Lucas: Yeah, that's the challenge. A bridge constantly moves from traffic, wind, temperature changes. So you need algorithms to filter out the normal behavior and flag anomalies. For example, if a strain gauge shows a beam is bending two millimeters more than usual under the same traffic load, that's a signal that something's changing. Luna: And you can catch that early enough to do something about it? Lucas: That's the promise. One of the most advanced examples is New York's Kosciuszko Bridge, which opened in 2019. They embedded over three hundred sensors into the concrete and steel — things like temperature sensors, accelerometers, even sensors that measure the tension in the cables. Luna: I've driven on that bridge. It's huge. So they're monitoring it in real time? Lucas: Yes. The data goes to the New York State Department of Transportation. They can see if a cable is losing tension, if a bearing is wearing out, if a joint is expanding too much. It's like having a doctor monitoring your vitals constantly instead of just checking your pulse once a year. Luna: But most bridges aren't new. What about retrofitting existing structures? The Fern Hollow Bridge was fifty years old. Lucas: Retrofitting is the bigger opportunity. You can attach wireless sensors to existing steel and concrete without drilling or wiring. They're powered by batteries that can last five to ten years, or even by tiny solar panels. Several companies — like Lord Sensing or Campbell Scientific — make these ruggedized units specifically for infrastructure. Luna: And how much does it cost to outfit a typical bridge? Lucas: It depends on the bridge size and number of sensors, but a study by the American Society of Civil Engineers estimated it's about fifty thousand to two hundred thousand dollars for a medium-sized bridge. That sounds like a lot, but compare it to the cost of a major repair after a collapse — or worse, a tragedy. The Fern Hollow replacement alone cost over two hundred million. Luna: So the return on investment is pretty clear, at least for critical bridges. Lucas: Absolutely. And the data doesn't just warn about imminent failure. It also helps with predictive maintenance — so you can schedule repairs when they're cheapest and least disruptive, rather than waiting for an emergency. Luna: That's the holy grail for infrastructure management. But are we actually seeing widespread adoption? Or is it still mostly pilot projects? Lucas: It's growing, but slowly. The U.S. has over six hundred thousand bridges — and about forty-two thousand of them are classified as structurally deficient. So the need is enormous. But funding is the bottleneck. Most DOTs rely on federal and state budgets that are already stretched thin. Luna: What about other countries? Are they ahead of us? Lucas: Japan and South Korea are probably the leaders. Japan especially, because they have so many bridges in earthquake zones. They've been deploying sensor networks since the 1990s. In Europe, countries like Switzerland and Germany have extensive monitoring on their highway bridges. The U.S. is catching up, but we have more ground to cover. Luna: One thing I've wondered: could these sensors have actually prevented the Fern Hollow collapse? Or would the warning have come too late? Lucas: That's a great question. The NTSB report cited severe corrosion in a steel tie that had been hidden by paint. A strain gauge on that tie would have shown abnormal stress accumulation over months or years. So yes — it's very likely that early monitoring would have caught it. The same goes for the 2007 I-35W bridge collapse in Minneapolis. That one killed thirteen people. Luna: It's hard to argue against that. So what's the next step? Are there any new technologies on the horizon? Lucas: The big trend is combining sensor data with AI. Instead of just setting static thresholds, machine learning models can learn the unique 'signature' of a bridge and detect subtle changes that a human engineer might miss. Also, there's work on using fiber optic cables that can measure strain along their entire length — essentially turning the bridge itself into a sensor. Luna: That sounds like something out of science fiction. But it's real. Lucas: It is very real. And it's already being tested on a few bridges in Europe. The key challenge is making the data actionable. You can collect terabytes of vibration data, but if it doesn't lead to a decision, it's just noise. The best systems are the ones that say: 'This specific bolt needs tightening in the next three months.' Luna: And that kind of specificity is what gets infrastructure managers to actually act. Lucas: Exactly. Look, this technology is not a silver bullet. It won't fix every corroded beam overnight. But it gives us a fighting chance to prioritize repairs and prevent disasters. And given how many of our bridges are aging — the average U.S. bridge is over forty years old — we need every tool we can get. Luna: It's one of those rare cases where the technology is proven, the cost is reasonable, and the benefit is literally saving lives. So why isn't it everywhere? Lucas: Inertia, mainly. And funding. But also, there's a cultural shift needed. Engineers have relied on visual inspections for a century. Trusting a sensor network is a leap. It takes time. Luna: That's a really important point. I think the same could be said for a lot of IoT applications — the tech is ready, but the institutions aren't. Lucas: Yeah. And that's why conversations like this matter. The more people understand what's possible, the more pressure there is to adopt it. Luna: Speaking of which, if today's conversation gave you something useful — maybe a new way to think about bridges, or a reason to ask your local officials about monitoring — that's exactly why we do these shows. We keep this podcast ad-free on purpose, because we don't want any sponsor to influence what we cover. If that kind of independence is valuable to you, you can support the show at buy me a coffee dot com slash fexingo. No pressure, just a way to keep it going. Lucas: Yeah, and it really does make a difference. Every bit of listener support helps us stay focused on the actual technology, not the advertising. Luna: Alright, back to bridges. One last thing — are there any surprising applications of this sensor technology that you've come across? Lucas: Actually, yes. Some researchers are using the same sensors to monitor bridges for scour — that's when flowing water erodes the riverbed around a bridge's foundations. It's a leading cause of bridge collapse in the US, and it's almost invisible from above. Acoustic sensors can detect the sound of sediment being carried away. It's fascinating. Luna: So the same sensor that detects a crack can also listen to the river eating away at the foundation. Lucas: Exactly. That's the power of IoT — one network, many threats. And we're only scratching the surface. Luna: Great episode, Lucas. I'll never look at a bridge the same way again. Lucas: Same here. Thanks for listening.