Fern Hollow Bridge should have been closed years before it collapsed
practical.engineering
practical.engineering
16 of the worst are in LA county and several see 300k trips daily, including one carrying the 405.[1]
0. PDF https://artbabridgereport.org/reports/2023-ARTBA-Bridge-Repo...
1. https://artbabridgereport.org/state/ranking/top-bridges
EDIT: States with the least % of SD bridges: AZ, NV, TX, DE, and UT.
WV and IA have the most at almost 20% SD bridges respectively. (1 in 5!)
EDIT2: Raw data https://www.fhwa.dot.gov/bridge/nbi/ascii.cfm
You need some form of a "stop ship" where inspectors get some number of "no way, close this entirely" that they can use without repercussions or something.
I feel like this thought process leads to the "worst possible category" containing 42k bridges. The construction of the categories was based around whether the bridge is safe, and this category says "the bridge is not safe." It's when you put it off long enough, and let the infrastructure deteriorate long enough, that you start going "which is the most unsafe though?" Or, "is a safety factor even necessary? They are by definition a >1.0 factor applied to what we think is safe." Then, "one in a hundred year events" or "once in a decade events." Eventually it's "hasn't collapsed yet!"
The fact that the list got that big means our problem isn't prioritization, our problem is failing to repair bridges. Commit to and budget for repairing 42k bridges, then prioritize where you start.
We need to hold road infrastructure to the same consideration as we do transit and rail. If we didn’t give it unlimited funding, then it would already be doing this.
We are wasting trillions and losing so much economic growth by doubling down on a bad investment year after year for decades. Imagine if this money was put into education or healthcare or transit. We’d actually have a net benefit!
Based on the evidence, it's hard to imagine that lack of money is the problem with those three things.
K-12 education in particular has seen more and more money spent over the decades while the number of students, teachers, and student outcomes have stayed more or less constant (the extra money mostly goes to administrators to administrate the extra money). Higher ed has gotten famously more expensive while delivering a poorer product, with grade inflation and recently a focus on worthless social activism diluting what it means to have a college degree at all.
On healthcare, the US already spends close to one-fifth of its GDP on it. Its per capita spend is over 50% higher than the next-highest country, Switzerland. And what do we have to show for this, other than an obesity rate exceeding one-third the population and diminishing life expectancies? I don't know how more money could possibly help this situation.
Transit only makes sense in a few cities (largely due to idiotic zoning that prevents the kind of densities that would enable transit) and even then, if you look at things like NYC's East Side Access project, it would be hard-pressed for a public works project to spend taxpayer money less efficiently. And to truly improve the transit experience would require a concomitant improvement in the social fabric, which is its own set of problems.
Local roads often don't qualify, but state and federal highways do.
And it can be done in a way that there isn't a bus on the bridge when it collapses.
So many ways to dice this information and I don’t even live in the US.
https://www.statista.com/statistics/190386/number-of-road-br...
The population of Texas is around 25.1 million people compared to 11.5 million in Ohio, a difference of 13.6 million people. That means about 2 times population wise.
I don't know by which size metric it is that ohio is 41 times smaller than texas?
Really? That does not comport with my experience growing up there.
The climate in Houston differs from that in Arlington, for example.
?
I'm not sure about anyone else, but I'm pretty sure our bridges should be something considered important enough to keep maintained. :(
> 36 percent of all U.S. bridges (over 222,000 spans) require major repair work or replacement. Placed end‐to‐end, these structures span over 6,100 miles – and would take over 110 hours to cross at an average speed of 55‐miles‐per‐hour.
That's a lot of bridge.
About ARTBA
The Washington, D.C.-based American Road & Transportation Builders Association (ARTBA) is a non-partisan federation whose primary goal is to aggressively grow and protect transportation infrastructure investment
https://www.federalregister.gov/documents/2017/01/18/2017-00...
Bridges (and buildings) are designed as systems. Water management is one of the biggest headaches in civil engineering - poor performance of any subsystem with a water management role is a legitimate cause for concern.
Kinda makes me think George Carlin was on to something with his hate of soft, euphemistic language.
My point is the report's summary could have said "This bridge will fall down within 3 years" instead of obscuring the with vague, watered-down jargon like "structurally deficient"
They finally passed an infrastructure bill which hopefully includes fixing 'em.
I know they've been replacing a lot on I-80 in IA.
Just like this bridge, the big problem is that something happens and degradation switches from "slow but manageable" to "Oh, shit, suddenly that beam is gone and this bridge might collapse".
While our state representatives are issuing a tax refund because we are "running a surplus"
This state is wild. I bet it will be the first to be merged with a nearby state because of poor governance.
Surprisingly, or not so, no one was found guilty, not even the inspectors that didn't report the ongoing damages, just because "it would be too costly to rebuild it, and profits of the highway company (1) in charge of it would be zero".
(1) the company name is Atlantia, fully owned at the time by the Benetton family, yes those of the sweater chain
There's even inspection reports of one of the cross supports rusted through and disconnected, a cable taking some of the strain. At the very least the bridge should have just been closed, as unsafe.
And the article starts by stating that the collapse was ‘without warning.’
When we do design, we're generally designing things so that if it fails it does so in a way that doesn't suddenly all collapse at once, but first deforms for a while so people can get away.
Edit: It was eventually overturned [1]. Still, a shocking indictment of the court system.
1. https://www.theverge.com/2014/11/11/7193391/italy-judges-cle...
If there were strong indications of cancer and a doctor blatantly ignored it? An argument could be made that's negligence causing death.
Clearly some degree of negligence is absolutely criminal when people's lives are on the line. Maybe you don't agree where that line was drawn in this specific case, but that's not an argument that no such line exists.
Obviously, since he isn't an Italian at all :-/
I'm asking because I've been in standstil traffic (in my car) on the Lisbon "golden gate" type bridge and that thing was freaking bouncing up and down from traffic on other lanes.
And yes, it's normal. That bridge is very well maintained (there's a whole team there working daily, afaik, and I've witnessed many interventions in the past).
Railways work well in the center of bridges. One bridge that violates this principle is the Manhattan Bridge; it was built before we had fancy engineering simulations or much experience with railway suspension bridges at all, and so the lower deck is a pair of railway tracks on either side of a roadway.
This unusual design is very stressful on the bridge and has resulted in expensive rehabilitation programs.
Floating bridges also don't necessarily need to have fixed points to load and unload. With suitable stabilisation and propulsion, they should be able to move to any number of loading and unloading points. We could have a floating bridge per train, vehicle, or any unit or set of users.
it was supposed to open earlier, but the contractor messed up the concrete pour and is fixing it at their own cost. but that doesn't have much to do with the fundamentals of the floating bridge itself.
But it's true that big structures are often designed to move as a way to deal with forces. It's much better than something firm that doesn't move until it snaps.
More to the point a proper study a year before the collapse with endoscopes showed the cables were pretty rusted and dangerous https://www.lastampa.it/esteri/la-stampa-in-english/2018/08/...
The company managing it had tendered to get repairs done but they hadn't started by the time of the collapse.
Norway has to be among the worst.
It is not fun and glorious for a political administration to set aside $$$$ every year that will just go to people doing boring work that the voters will not be impressed with.
You dont see politicians "Under my administration we painted XX buildings, we did need maintenance of YY bridges, we replaced ZZ parts of the railway that would become problematic with time.
Rather: "Under my administration we opened up a new large hospital (because the other had near 0 maintannce for decades), we built 2 new bridges etc"
That's just bad politicking then.
"Under my administration we hired hundreds of workers in Anytown who can proudly say they worked to maintain this city's infrastructure and provide a future for not only their kids but all of our kids. Their paychecks put food on the table and money into the local economy."
Apparently, nobody pays for it either way. It's astounding to me how little tax money goes into paying for infrastructure.
In 2023 the US federal government spent $44.8 billion on infrastructure and transferred $81.5 billion to the states. That's $126.3 billion out of a $6.1 trillion budget. ('Only' $4.4 trillion in revenue though.)
That's 2% of the budget.
> The federal government spent $44.8 billion on infrastructure in 2023 and transferred an additional $81.5 billion to states. In 2021, state and local governments spent $218.5 billion on transportation and infrastructure, excluding federal government transfers.
[1] https://usafacts.org/state-of-the-union/transportation-infra...
Also note (a few rows down the table)[2] that the US government spends more than it collects in taxes. In 2022 the expenditures added up to $9.64 trillion.
Sure, it's not 2%. It's 3.6% instead.
[0] https://www.federalreserve.gov/releases/z1/20240607/html/f10...
[1] https://www.federalreserve.gov/releases/z1/20240607/html/f10...
[2] https://www.federalreserve.gov/releases/z1/20240607/html/f10...
Having localities pay from local tax money seems like it would focus money on expenditures that the people that use them and pay for them will approve of with their own skin in the game. There are some interstate highways that serve multiple localities, but these should be the minority, right?
We should cover all our internal needs before we show our largesse elsewhere.
And, like maintenance: foreign aid can avoid local problems, e.g. stabilizing countries in central America can reduce the incentive for people to flee to the US (which for most people is a distant plan C over staying where they are or moving a short distance away).
That money could be spent on maintenance rather than fattening our MIC.
NGOs are similar. In theory they sound good, the leaders can pull sizable salaries, eclipsing any congresspersons earnings and they are incentivized to keep the bad thing happening—if they solved the problem their reason to be would cease.
The reason is geopolitical. It is seen as necessary to give money to allies to prop up their military so the US hegemony can be maintained. Adversaries (China, Russia, Iran) need to be made afraid not only of US might, but also of Allies’ might. So the US doesn’t wait around for those countries to spend on their military on their own, or god forbid to realign their alliances based on their spending. Checks it is.
But contrary to popular belief, the MIC isn’t that fat. They are mostly public companies - you can check out their profits yourself. Here’s one: https://valustox.com/NOC
Their margins tend to be small. Their revenues are incredibly steady (they’re pretty much government departments). Staying ready for war like this is actually a wise course of action - it would be catastrophic to have to ramp up in the next emergency like in WW2.
The military is in desperate need of more new contractors like Anduril and SpaceX to provide competition for the incumbents, as well as stronger collaboration with allies like having Japanese shipyards build some of our ships.
The programs and systems may not be perfect but at least there are programs and systems.
I think the mockery was over how he transactionally framed it rather than the principle itself. But TBH do we really want European countries to rearm? After centuries of fighting they quieted down after outsourcing it to the US. And since they weren't fighting that outsourcing wasn't even that expensive. I am sure it was cheaper to have all those US troops supporting NATO countries than to get drawn into yet another war over there.
> Staying ready for war like this is actually a wise course of action - it would be catastrophic to have to ramp up in the next emergency like in WW2.
It's like insurance -- you hope that it's a deadweight loss but pay for it because it's cheaper than holding the risk yourself. And I do think the western countries overdid it in regards to downsizing after the end of the cold war.
Since most income and business taxes go to the Federal government, states are dependent on Federal grants for a lot of infrastructure.
That strange to hear. Why do you have this feeling? After visiting Porto and Italy around and below Napoli, I cannot imagine that there are any place in Nordic countries which has even just similar tolerance to not maintain something. But I don’t know too much about Norway specifically.
I think the bigger problem is maintenance is just one cost out of many. There's also education, and health care, and social services, and police, and firemen, and pensions, and all sorts of other things, and that's also important. It's relatively easy to "save" on maintenance because nothing is going to fall down immediately and no one will really notice – at least not for a while. In the long run you're not really saving anything of course.
It's easy to critique this from the sidelines, but the pressures politicians and governments are under make it pretty tricky to do anything else. Saving money in other areas is going to be unpopular. Raising taxes even more so. A lot of times stuff like this is a Kobayashi Maru.
Let's say it would collapse with 100% certainty randomly in the next three years, and you're in the danger area for 2 minutes, with a 20% chance of fatality (in fact, nobody died). That's still around a one in 4 million chance any given trip kills you, about the same as 30 miles of driving for the average American driver.
Most people would accept that level of risk. Perhaps not to save a couple of minutes on the journey, but if everyone was redirected to another route at rush hour, it might cost each commuter 10-20 minutes.
A handful of newsworthy bridge collapses per decade across the US doesn't seem so bad. Instead of negligence, perhaps that indicates an appropriate level of maintenance and risk tolerance, and an appropriate human price to add to the 500,000 other road deaths over the same period.
Case in point right now at the France-Italy border : they're building the longest rail tunnel in the world under the Alps for tens of billions. There is a tunnel already (Frejus tunnel), which is used at 25% of its capacity, but public money is raining on this unstoppable, useless project.
On the other end, all over Europe and particularly in France and Italy, rail and road infrastructures are in a state of disrepair. Wouldn't this money be better used repairing thousands of bridges, tunnels, etc? Sure, but no politician could campaign on "I've got billions spent in our region!". No multinational civil engineering and construction company such as Eiffage or Bouygues would get big money. All the big boys are in favour of the project, and all the small guys simply don't count.
This is all completely fucked-up, frankly. It's a general failure of democracy and a global corporate takeover.
https://en.wikipedia.org/wiki/Turin%E2%80%93Lyon_high-speed_...
Additional traffic limitations stem from the impact of excessive train transit on the population living near the line. Some 60,000 people live within 250 m (820 ft) of the historical line, and would object to the noise from late-night transits. In 2007 the conventional line was used for only one-third of this calculated total capacity. This low use level was in part because restrictions such as an unusually low maximum allowable train height and the very steep gradients (26–30‰) and sharp curves in its high valley sections discourage its use.
A 2018 analysis, by contrast, found the existing line close to saturation, largely because safety regulations now prohibit passenger and freight trains from crossing in a single-tube tunnel. This very significantly reduces the maximum allowed capacity of the 13.7 km long Fréjus tunnel, which trains of one type must now fully cross before any train of the other type can be allowed in the other direction. The historical line's path through the deep Maurienne valley is also exposed to rockfalls, and a major landslide in August 2023 forced its closure for most likely over one year.
With the amount of money invested you could probably fly everyone from Lyon to Turin on expensive aviation biofuel for centuries before recouping the humongous investment. This is just stupid grandiosity allied with corruption.
In recent years there have been plans to "boulevardize" some major arterials within city boundaries, meaning the city takes over responsibility of the road segment and turns it into a street with lower speed limits, multimodal access, and so on. The intention is, of course, to facilitate urban development in the corridor.
The core network of major roads in Finland can be kept in a good shape, the issue is largely the thousands of kms of secondary and tertiary roads in the middle of nowhere that have too little traffic to be any sort of a maintenance priority, or to be eligible for EU funding. But you can’t really help it in a country that’s as sparsely populated as Finland outside major urban areas.
Also I don't think it's possible for your risk calculation to be done preemptively. We don't know the final breaking point of a piece of steel until we break it. All the calculations and modelling of the bridge will have been done with error margins and because of that we don't have any choice but to over-engineer things so we always stay outside of the worse-case margin for error. Given all that, I don't think it's unreasonable to expect the risk of a bridge collapse to be a lot lower than the average risk we take on the road.
For those wondering if that is right, it is. Here's the math.
Americans drive about 3.2 trillion miles per year and about 40 000 people are killed. That's one death per 80 000 000 miles.
Assuming each mile is equally deadly, that chances you survive a given a mile would be 79 999 999 / 80 000 000. To survive a trip of N miles, you have to survive each individual mile sequentially. The chances of that would be (79 999 999 / 80 000 000)^N.
The chance of not surviving that trip would then be 1 - (79 999 999 / 80 000 000)^N.
For N = 30 that is 1 / 2 666 667, which is close enough to dmurray's number to count as a match. There's enough fuzziness is in the inputs that all we can hope for is the same ballpark.
I've seen others say the rate is one death per 120 000 000 miles, and for N = 30 that does give 1 / 4 000 000, so I'd guess they are using that rate.
> Most people would accept that level of risk. Perhaps not to save a couple of minutes on the journey, but if everyone was redirected to another route at rush hour, it might cost each commuter 10-20 minutes
One big difference is that with the bridge everyone has the average risk. I cross the bridge, I'm rolling a d4000000 and hoping I don't get a 1.
With a car I can take steps to make the chances of dying on my particular trip much lower than average. With the car I can often time my trip so as to go at times of day or during weather conditions or during traffic conditions when accident rates are lower.
Or how about let's say it would collapse with 100% certainty randomly in the next three years, and a school bus with twenty kids in it drives over the bridge twice a day, with each child facing a 20% chance of fatality.
‘I probably won’t be the one who dies when it collapses’ is a terrible metric for whether or not we should try to mitigate the risk of a bridge collapse.
Bridges sometimes fail catastrophically, with risk to life. Modeling risk is a necessary way to consider the costs of mitigation.
Leaving aside the poor helpless babies, what metric would you suggest?
I wouldn't and I definitely wouldn't if I had other people in the car.
:)
They're not really equivalent, unless you've got a high-risk condition like asthma.
Statistically speaking, a 100% chance of losing one year of lifespan (out of the 76 years an American can expect) is a greater mortality risk than a 1% chance of instant, immediate death.
In fact, if you're aged 26, a 1% chance instant death has half the mortality risk - and the older you get, the better it looks.
However, if you look at humans' revealed preferences, people would much rather lose one year of lifespan than take a 1% chance of losing their entire remaining lifespan.
Mortality risk isn't a very good predictor of human preferences, it turns out.
Statistically speaking, a small chance of losing everything today is actually worse than a guaranteed chance of losing a smaller amount later.
I am fairly certain that a large majority of people, if they'd have had access to those images of the bridges structural members, would have stopped using it.
I think most people can accurately diagnose that as an actual, real problem.
It's an important enough member to warrant 10 (I am assuming symmetry here, you cannot see all of them) not quite tiny bolts and nuts, but you can see clearly through the thing.
https://gis.penndot.gov/paprojects/BridgeConditionsMap.aspx
... and find some that pass over other roads, and you can find other instances:
https://www.google.com/maps/@40.4562613,-79.9736136,3a,52.6y...
...but not for the reason you're implying.
The "rust belt" used to be the "steel belt", until free trade agreements made the US steel industry non-viable.
Rust is symbolic of that decay.
All of the infrastructure that was built when these areas were economically solid, is now rotting along with the economy that built them.
How does that 1:4M probability change when you commute over the bridge twice a day? Humans are notorious for being bad gauging risk, and we very often make the mistake that past success means lower risk of future failure, which for a deteriorating bridge is exactly the opposite of what's going to happen.
In my opinion bridges shouldn't collapse, like ever. Annual inspections for years prior pointed to a severely deteriorating structure, even after the temporary cable stays were put in place. The tweet with a picture of a completely detached member a couple of years before the collapse makes this even more egregious.
This is a textbook example of where bureaucracy prevails over common sense. Heads should roll. Thankfully nobody died, but the lack of maintenance and upkeep resulted in a total failure, wasting lots of taxpayer money to replace.
You can get a good sense of this from the equivalence listed immediately after the odds:
>> That's still around a one in 4 million chance any given trip kills you, about the same as 30 miles of driving
So it'd be closely analogous to driving a 30-mile commute twice a day. How much risk do you feel that involves?
Moreover, people generally perceive things in which they (theoretically) have more control over as safer - You can control/mitigate the risk you undertake as a driver to some extent (drive slower, bigger gaps, etc), not so for a random bridge collapse.
Seems possible. Airline booking sites already offer filters to exclude 737 Max planes.
> That's still around a one in 4 million chance any given trip kills you
That's a one in five chance that any given trip kills you, unless 20% means something very different to you than it does to me.
Collapse of the Fern Hollow Bridge Pittsburgh, Pennsylvania January 28, 2022
Highway Investigation Report HIR-24-02 released: February 21, 2024
PDF (136 pages): https://www.ntsb.gov/investigations/AccidentReports/Reports/...
Which bridge that is currently in operation should be closed next?
(Not a dunk on the article, which brilliantly addresses the difficulty of knowing in advance vs making real world changes. Practical Engineering is an awesome YouTube channel!)
The article/video actually touches on this:
> The City of Pittsburgh quadrupled their spending on inspection, maintenance, and repairs. And they redid the load ratings on all the bridges they owned, resulting in one bridge being closed until it can be rehabilitated and two more having lane restrictions imposed.
I don't know which one bridge it is, though.
Very thankful that nobody was hurt when it collapsed, and as other people have pointed out it is representative of all of the infrastructure that many cities have but can no longer afford to maintain or replace.
https://www.cbsnews.com/pittsburgh/news/completely-preventab...
But it's still not a very good comparison, because airplanes, due to their very nature, can't be designed to fail safe. Nuclear power plants can.
maybe they should. maybe we could celebrate repair like we do new construction. there’s a comfort in knowing we’ve been put good again that’s worth signifying.
But to the bigger point made near the end, without a person in the loop who both appreciates the meaning held within the inspection reports AND having the power to act on that information, we still remain vulnerable to the complexity of our own social systems becoming too inefficient to handle problems like this.
That's why ridiculously common case summarized as "pilot error" usually involves several components including organization, training, etc.
EDIT (After reading the report): And indeed, "what we found" section and "what we recommend" is all about how PennDOT and related orgs operate.
"...collapsed without warning. ...And this bridge had been listed as being in ‘poor condition’ for over a decade. "
Bridges and overpasses that exist until they break and the people die. People call out to God for justice! But all that infra is then rebuilt the same way if it’s rebuilt / when it’s rebuilt.
Something something American tech people have forgotten just how amazing the US is because they don’t realize how good it is. They take too much for granted. They have the safety to be snarky on the internet.
Get a passport and go for a walk.
Why not compare to country where bridges dont fall at all?
On the engineering side - you learn far more from analyzing failures than from analyzing successes.
On the social side - there's nothing mysterious about how the US bureaucracy failed here. Briefly contrasting that with Utopialand (where the society & government are different, and bridges never fall) can work as journalism. Or as a rebuttal to "failures will always happen" doomsayers. But the utility is pretty limited. The US isn't a tech company, where you might fire up the troops by talking about how your competition is delivering obviously-better results on metric X.
This is just sophistry. You won't learn how this collapse could have been prevented, or how to prevent others like it, by studying countries where infrastructure is worse. You're already doing better than those places, and still it's not enough.
Also, I'm pretty sure engineering schools study both failures and successes. It is incredible to me that someone would honestly believe studying bridges that have not fallen is useless.
Another engineering quip: "Any idiot can build a bridge that stands. But it takes an engineer to build a bridge that barely stands."
There's a wide range of circumstances there, which makes sense since "third world" was a political designation, not an economic one.
Do you also complain when people use the term "decimated" and they aren't talking about killing 1 of every 10 soldiers?
And then there's a political commentator who calls the United States (and the United Kingdom) a Fourth World -- and that's not a positive thing. https://archive.is/UIr49
How America Collapsed and Became a Fourth World Country
How America (and Britain) Became Failed States