The Bridge Collapse in Miami

It goes without saying that this is a disaster:

At least four people have been killed and 10 others hurt after a footbridge collapsed near Florida International University in Miami, officials say.

The 862-tonne, 174ft (53m) bridge fell over an eight-lane motorway on Thursday afternoon, crushing at least eight vehicles, police said.

They told local media that vehicles were stopped at a traffic light when the structure collapsed at about 13:30 local time (17:30 GMT).

It is still unclear how many people were under the bridge at the time.

For a standard, single-span footbridge to suddenly collapse in this manner in the United States in 2018 is incredible. Reinforced concrete footbridges have been built worldwide for decades, and ought to be the sort of thing a highways department can design and build on their own. Perhaps one 53m long over an 8-lane highway requires some specialist assistance, but still. This was not the Millau Viaduct.

These days engineering firms use finite element modelling (FEM) programs which can predict how a structure will perform under load, identify any weak points, and calculate the stresses induced in every location. The bridge which collapsed looks to be of an unusual design, no doubt approved by an architect somewhere, but that is precisely what the FEM models are for. Using them you can depart from a traditional design, incorporate architectural features, and still be sure your structure is sound. A newly installed footbridge suddenly collapsing onto traffic in the US is the equivalent of a batch of canned goods killing people, the contents having reacted to the metal. There is enough knowledge and experience by now to ensure these sort of accidents no longer occur.

Yet it did, so why? The BBC tells us this:

The bridge was erected on Saturday in just six hours.

It was built using a method called “accelerated bridge construction” to avoid traffic disruption. A major section of the bridge was assembled on the side of the road and then raised into place.

I’ve written before about bridge collapses, and how they tend to happen during construction, but I don’t think the installation method described above had anything to do with it. There’s a video here of the bridge being installed, and it looks to me like a pretty common technique which doesn’t in itself explain why it collapsed a few days later.

So here’s my guess: somebody screwed up the calculations or the finite element model, and nobody picked up the error. If this was in the developing world I’d be more inclined to believe it was shoddy construction or poor materials rather than a design error, but this being the US I can’t see that happening. The company that carried out the engineering was Munilla Construction company, a family-owned firm whose website is here. They’re based in Miama and have been around since 1983 and they claim:

WE BUILD YOUR PROJECT ON TIME, ON BUDGET, AS DESIGNED AND WITHOUT ANY SURPRISES.

They might want to update that at some point.

What follows is pure speculation on my part but it’s based on what I’ve observed of engineering companies and corporations over the past 15-20 years or so. Back in 1983 the firm would have employed serious engineers who held themselves accountable, and the brothers who founded it probably did a lot of the work themselves. There is no doubt this company was able to successfully deliver engineering projects for a couple of decades, so they’re not some fly-by-night outfit owned by the wife of the local mayor.

But I suspect things changed sometime in the past 5-10 years. There is nothing on the company website to suggest they succumbed to the relentless pressure placed upon firms to hire people based on their appearance and sex rather than competence and ability, but I’d be surprised if they were wholly unaffected by demands for greater workforce “diversity”. After all, this was a firm which did several projects for the public sector, and installed the bridge on behalf of a university. It’s unlikely they weren’t required to demonstrate they were fully on-board with the latest progressive directives. At a guess, I would say this is a company which has seen several experienced engineers retire over the past decade, replaced with people whose abilities are questionable.

Secondly, as I have complained about in the past, there has been a major shift in modern companies from delivering something useful – such as a bridge which doesn’t collapse – to managing processes. A lot of companies have subcontracted out the actual work – designing, building, manufacturing, operating, maintaining – and instead busy themselves with “managing” the whole process. This involves lots of well-educated people in nice clothes sitting in glass-fronted office buildings sharing spreadsheets, reports, and PowerPoint presentations by email and holding lengthy meetings during which they convince one another of how essential they are. I’m sure this is pretty much what Carillion was doing when they went bust: anything useful was done by subcontractors. The distance between those doing the actual work and those supposedly responsible for the outcome has, in far too many companies, grown into a yawning chasm. Survival in a modern company is all about compliance and obedience, and accountability is non-existent because it is no longer required.

In such an environment, it is inevitable that the quality of work suffers, errors go unnoticed, and – occasionally – catastrophes occur. Now I don’t know if that was the case at the Munilla Construction company, but somehow they’ve gone from an outfit who could deliver a project with their eyes closed to one that has just dropped a simple footbridge on eight lanes of highway. If I were investigating, I’d want to know who did the actual design and where it was done. I’d be willing to bet a hundred quid the calculations and finite element modelling were done outside the US to save money, or subcontracted to another company, and supervision – which involves expensive Americans – was at nowhere near the levels it should have been. Regardless of where they were done, I’d also be willing to bet the company spent more manhours on progress meetings and overly-detailed weekly reports to let the management know what was going on than they did checking the engineering calculations.

I might be wrong, and maybe I’m being unfair to the Munilla Construction company. But I’m not wrong in describing how modern companies work. I have a hunch we’re going to see more disasters like this in the coming years as successive generations of managers and engineers fail to deliver what were, until recently, pretty ordinary projects.

UPDATE

David Moore posts a link to this video which reckons the bridge was supposed to be supported from above using cables in the final design. If they’re right, then this is a classic case of a poorly-supported bridge span collapsing during construction, and we can all go and read my previous post. It doesn’t make the cock-up any less severe though, just slightly more understandable.

Liked it? Take a second to support Tim Newman on Patreon!
Share

61 thoughts on “The Bridge Collapse in Miami

  1. I kind of doubt this one is cause by the usual corporate drive to institutional uselessness.

    There is serious shortage of good engineers around the world, the ones who understand an entire system and how it goes together. It doesn’t take much staff turnover to see some huge gaps in knowledge open up.

    Although there is this comment;

    “a state-of-the-art bridge made with new, high-tech materials.”

    Hmmm, new materials…..

  2. There is serious shortage of good engineers around the world, the ones who understand an entire system and how it goes together. It doesn’t take much staff turnover to see some huge gaps in knowledge open up.

    Is this not a symptom of institutional uselessness? If good engineering and accountability is not rewarded in an organisation, it will disappear. If arse-licking is rewarded, that’s all you’ll be left with. Incentives matter, as Mr Worstall likes to remind us.

  3. Hmmm, new materials…..

    The design looks dodgy to me (what’s that bracing on the top supposed to be doing?) But then, a lot of the modern ones do – that’s what the FEM systems are for.

  4. I read a witness report that said a crane dropped a load onto the bridge, triggering the collapse – unconfirmed

  5. “Is this not a symptom of institutional uselessness? If good engineering and accountability is not rewarded in an organisation, it will disappear.”

    Agree that this is a major problem in the bigger engineering firms.

    “Incentives matter, as Mr Worstall likes to remind us.”

    Yes. I’ve just done a project management presentation that has the first three points as;
    1. Incentives matter
    2. Incentives really matter
    3. Incentives really, really matter.

    Apply that to this incident and we may find a false incentive there somewhere..

  6. To add to “incentives matter”, another important principle is that You Get What You Measure. With the corollary that whatever you don’t measure gets worse to compensate. And the proviso that what you measure might not be what you think you’re measuring.

  7. I read a witness report that said a crane dropped a load onto the bridge, triggering the collapse – unconfirmed

    If so:

    1. The crane shouldn’t drop its load.
    2. The crane should not be lifting a load that, if it drops, destroys the bridge.
    3. If you’re doing *anything* which has even an outside chance of destroying the bridge, stop the traffic and secure the area.

    A litany of fuckups.

  8. To add to “incentives matter”, another important principle is that You Get What You Measure.

    The problem with that is modern managers now want absolutely everything measured, and they’re swamped with data. A weekly report on a project in the oil industry runs to around 80-100 pages. I was once on a course on Engineering Management an the lecturer said there are around 3-5 metrics which tell you if you’re on track or not, the rest is just noise.

  9. @Tim – indeed. I won’t repeat my earlier-recounted tales about previous employment where they had a cycle of “identify metric. Measure for info. Insist it won’t become a target. Remark difference between sites. Make it a target”.

    The only 2 metrics that worry me now are “are we respecting all deadlines” and “are we making enough money to function comfortably”.

  10. abacab

    “To add to “incentives matter”, another important principle is that You Get What You Measure. With the corollary that whatever you don’t measure gets worse to compensate. And the proviso that what you measure might not be what you think you’re measuring.”

    I strongly disagree! I’ve simply seen too many companies try to measure everything and then utterly lose control of the whole operation. Everyone starts working to the ‘measures’, and barely a sole looks at the actual goal. It perverts the entire business.

    Edwards Deming showed this decades ago.

  11. I strongly disagree! I’ve simply seen too many companies try to measure everything and then utterly lose control of the whole operation. Everyone starts working to the ‘measures’, and barely a sole looks at the actual goal. It perverts the entire business.

    But that’s a perfect example of the phenomenon – the actual goal is not measured. So gets worse to compensate everyone working to the measures.

  12. abacab, I’ve completely misread what you wrote, so I revise my comment to strongly agree!

  13. The excellent AVE has put up a video on the bridge, good watch;

    Awesome! This changes things somewhat. Post updated.

  14. Incidentally, the video refers to a bridge that collapsed during construction in British Columbia. They retrieved the metal from that bridge and turned it into rings, which are now given to chartered engineers in Canada and worn on the little finger of the hand they draft with. It’s how you can spot a Canadian engineer.

  15. The amount of experience that I’ve seen just walking out of the door of major oil companies in the last few years has been frightening, in he main due to their being no effective management of change in the “succession planning” (ha!). The incoming replacement is more often than not someone who has been fast tracked as part of a nationalisation programme and I would say, generously, that 25% might be ready to take up the post – usually as they had been mentored for the preceding three years. The majority are dropped in post with a hasty handover to a) tick the target box for the local content quotas & b) to slash the wages bill ASAP.

    Having seen this in projects and construction, exploration drilling (where the problems Tim describes have been carried across into things as critical as high pressure well design) and operations, I’m sad to say that a whole lot of stuff might be going ‘pop’ and ‘bang’ in the next decade.

  16. You sound a bit like David Lammy on Grenfell here. Don’t know what’s happened, but already know who’s to blame.

    It’s not unreasonable to observe that there are always human failings behind technical ones, but it would be a lot easier to identify the relevant humans *after* we know what the technical failings actually are.

    It might be Chinese re-enforcement…

  17. You sound a bit like David Lammy on Grenfell here. Don’t know what’s happened, but already know who’s to blame.

    Didn’t he blame the national government?

    It’s not unreasonable to observe that there are always human failings behind technical ones, but it would be a lot easier to identify the relevant humans *after* we know what the technical failings actually are.

    If those human failings don’t ultimately lie within the engineering company responsible for the design and construction of the bridge, I’d be interested to know where they do.

    It might be Chinese re-enforcement…

    This would support my case, not undermine it.

  18. I’m sad to say that a whole lot of stuff might be going ‘pop’ and ‘bang’ in the next decade.

    It’s high time they employed middle-aged Brits to prop up bars in exotic locations and be ready for when they do.

  19. IIRC, the “new material” was supposed to be “green” – some sort of cement additive (unobtanium oxide?) that reduced the need for cleaning. All my hasty search could find was a comment about it being “constructed entirely of self-cleaning concrete.”

    -XC

  20. “It’s high time they employed middle-aged Brits to prop up bars in exotic locations and be ready for when they do.”

    Well, quite.

  21. I’d guess failure of the concrete pre-tension system in the lower deck slab. Combined with the lack of structural support (that the overhead suspension was to eventually provide), it just wasn’t strong enough to hang there by itself.

    There’s a good video at https://www.liveleak.com/view?t=ssd8Q_1521180310 that shows the deck collapse. You can see the lower deck suddenly slump a bit as it breaks, and the tension cables slip and then the lower deck cracks open and it all comes down.

  22. James Elam: ’IIRC, the “new material” was supposed to be “green” – some sort of cement additive (unobtanium oxide?) that reduced the need for cleaning.’

    Green Schemes Kill.

  23. IIRC, the “new material” was supposed to be “green”

    Perhaps it’s the structural equivalent of the “green” cleaning products, that don’t clean, the “green” weedkiller that doesn’t clean weeds, or the “green” wood preserver that doesn’t preserve wood?

  24. Wow on that architectural model on the dudes video it shows the bridge as being cable stayed. There is also tie in points clearly shown on the top of the top section after collapse. I can’t understand why if it was a cable stayed design why they would have had no falsework in place until the bridge was hung from the cables.

  25. Self cleaning concrete is basically concrete with Titanium Dioxide powder dissolved into it when wet. It then serves to catalyse the breakdown of Nitrogen dioxide and exhaust fumes using UV light from the sun. It’s the same stuff that you find in road paint and isn’t remotely new as I did a project on it at A-Level. It doesn’t affect the strength of the concrete in any way and it seems sensible when you consider that that bridge appears to have traffic idling away underneath it.

    As for how exotic it is – You’ve had it it in your mouth….Find the back of a packet of toothpaste and it’s usually there in the ingredients list…

  26. I can’t understand why if it was a cable stayed design why they would have had no falsework in place until the bridge was hung from the cables.

    As the guy says in the video, they probably wanted to get the traffic moving. It’s understandable not to want the road blocked for very long, but somebody screwed up the calculations somewhere. The FEM analysis should have been applied to every stage of the construction process, not just the finished structure.

  27. It doesn’t affect the strength of the concrete in any way

    You’d certainly hope not, but when I learned the “green” cladding on the Grenfell Tower wasn’t fireproof…

  28. Mmmm. Seems like others think the same thing as you do.

    Oof! I bet she wishes she’d not said a word.

  29. “The FEM analysis should have been applied to every stage of the construction process, not just the finished structure.”

    It would have been, it just seems strange and kind of impossible that they would not have known that the bridge needed temporary support up until the cable column was erected and the cables tied along the span. I doubt they would have said well we can’t stop traffic until the cables are installed and we will just have to hope that it is okay until we install it. I read that they were doing some post tensioning on the lower flange when it collapsed, the only thing I can think of was that they had designed it such that this post tensioning could temporarily compensate until the cables were installed. Has anyone said where the cable column and cables were (ie stored nearby ready for installation) when the bridge collapsed?

    Unless of course it wasn’t designed as a cable stayed bridge and that architectural 3D modelling on the video that showed it as cable stayed was not an accurate reflection of the final design which was not to be cable stayed and just a simply supported span.

  30. Unless of course it wasn’t designed as a cable stayed bridge and that architectural 3D modelling on the video that showed it as cable stayed was not an accurate reflection of the final design which was not to be cable stayed and just a simply supported span.

    That is a real possibility. For all we know, that might have been something just knocked up for the bid to impress the client.

  31. We are living in an Atlas Shrugged world. Where stuff that previously was entirely within our technical capabilities suddenly starts to fail catastrophically, because the people in charge of it are either badly educated/have fake credentials, or owe their jobs to political ideology over competence, and/or have made decisions based on political ideology rather than science.

    In a world where no one has to accept responsibility for their actions, ever, this sort of thing will be happening more and more often.

  32. Although, on the video at 5.35 you can see in the photo on the top side of the top flange what looks like a tie in point (a plinth like concrete protrusion with integral steel ties for the cables) then further up the photo just at the very top you can see what looks like the start of another cable tie in point.

  33. I’ve lived in the U.S. for 58 years and I can easily imagine shoddy materials and sloppy work. New houses here are notorious for same.

    People keep voting to make this a tribal-type 3rd-world country. Less oversight, less accountability, more tribalism and cronyism. You can call your tribe “woke” or “progressive” or “conservative” or whatever you like, but a tribe is a tribe, and if you select for tribal membership rather than competence when you’re doing things like building a bridge you may not get very good results. At this point liberals will say “But the rich white guys of the 20th century were a tribe!” They were. And one of the rules of tribal membership in that particular tribe was that if you embarrassed the tribe too badly, like with your brand-new bridge collapsing, they would turn on you to protect the tribe as a whole. Today’s tribes aren’t like that.

    In short (finally)—if you vote for third-world conditions, whether because you don’t like white guys or because you don’t want to pay taxes or because whatever, you will get 3rd-world problems.

  34. The first thing I thought when I saw the bridge up was, why are the roof supports at funny angles? You would expect them to be symetrical. Then you see the picture with the cables and it all makes sense. What doesn’t make sense though is if it’s a suspension bridge why wouldn’t you erect the cable tower first?

  35. They retrieved the metal from that bridge and turned it into rings, which are now given to chartered engineers in Canada

    That’s actually a myth.

    Not to antagonize our host, but one of the reasons I got the hell out of engineering in Canada was the observation, initiated in undergrad and confirmed via twenty years of working with and for engineering companies, that the least trustworthy person in the room was the bloke with the stainless steel ring on his little finger.

  36. That’s actually a myth.

    Shit, I heard it from two Canadian engineers, separately!

    the least trustworthy person in the room was the bloke with the stainless steel ring on his little finger.

    Heh!

  37. “Incidentally, the video refers to a bridge that collapsed during construction in British Columbia. They retrieved the metal from that bridge and turned it into rings, which are now given to chartered engineers in Canada and worn on the little finger of the hand they draft with. It’s how you can spot a Canadian engineer.”
    Only partly correct. The rings are given to engineering graduates (I’m assuming the term ‘chartered’ is roughly equivalent to ‘Professional Engineer’ here in the colonies).

    A good summary is here:
    https://en.wikipedia.org/wiki/Iron_Ring#Ritual_of_the_Calling_of_an_Engineer

    The Kipling poem used in the ceremony:
    http://www.kiplingsociety.co.uk/poems_martha.htm
    https://en.wikipedia.org/wiki/The_Sons_of_Martha

    EDIT: Well, you’re definitely getting corrected on this, aren’t you?

    D.R.: You need to hang with a better crowd.

  38. Bruce Charlton has written about that.

    Yup, he beat me to it. It’s somewhat depressing to realise how far down this path of decline we are: I’d assumed it was a recent thing.

  39. D.R.: You need to hang with a better crowd.

    I’m assuming it’s regional. The US and Western Canadian engineers I’ve worked seemed like straight shooters. The Mac and U(W) grads I’ve worked with in the GTA were mendacious sleazeballs, to a man.

  40. Yep it’s cable stayed alright. On face value this looks like a monumental fuck up of the highest proportions, I honestly cant think of any plausible reason for putting the column and cables up last. As I said earlier unless they thought some kind of post-tensioning of the lower flange would be a temporary solution. This really does defy reality and some folk involved in planning its ill fated erection might get locked up for this one.

    “Miami bridge that collapsed lifted into place without suspension cables, support tower

    Cable-stayed bridges have cables attached directly from the column to the span, while suspension bridges string cables between towers and have other cables descend to the span.”

  41. Yep it’s cable stayed alright. On face value this looks like a monumental fuck up of the highest proportions, I honestly cant think of any plausible reason for putting the column and cables up last.

    What’s worrying me is all the news reports, supported by the postings on social media from the client, reference “the completed bridge” that was installed last Saturday. None of the news reports has said that this was a partial construction. So either the news reports are woefully wrong – which is quite possible – or the design underwent a pretty fundamental change late in the day.

  42. Rings ‘n’ things.
    I was an engineer , electronic, and many many years ago a mature fellow engineer advised me never to wear a ring.
    If working on moving stuff it will snag and you will end up digitally impaired, working on EHT it will provoke an arc, and working on low voltage it will cause a short and the finger will get v hot.
    Of course if you are a desk bound engineer wear all the jewelry and piercings you want. But heavy rings make keyboard operation more tiring.

  43. @Doonhammer

    The same can be said of not wearing gloves when working mechanically with your fingers, the initial touch sensitivity being removed can reduce your reaction time to such a degree that you lose your hand or worse if it doesn’t come off get entangled into a moving conveyor and lights off.

    By the way in my experience it never comes down to things like that. I have investigated many serious workplace injuries over the years. I was recently part of an investigation into a fatality on a construction site in Oman. There wasn’t that much to find, the poor chap was standing in the wrong place at the wrong time and paid the ultimate price. Most of my time was spent around coaching witnesses for a court appearance, getting his remains repatriated to his home country, negotiating a compensation payment figure for his family, making sure the payment was made to the wife and not a male relative and reassuring the expat managers that they were going to flung in prison.

  44. The “On the folly of rewarding A, while hoping for B” has been going on for a long time. This February 1995 essay sets it out:

    https://www.ou.edu/russell/UGcomp/Kerr.pdf

    If it is dated 1995, then there must have been a body of evidence and research based on that evidence collated and analysed prior to that. An absolute minimum of 25 years since this was becoming obvious, likely longer …

    Possibly it is linked to the introduction of MBA’s and the jump from newly qualified to MBA directly without any intervening experience and/or practical work in between.

Comments are closed.