Failing Infrastructure and the Lessons It Teaches

Infrastructure seems so permanent and mundane that most of us never give it a second thought. Maintenance doesn’t make for a flashy news story, but you will frequently find a nagging story on the inside pages of the news cycle discussing the slowly degrading, crumbling infrastructure in the United States.

If not given proper attention, it’s easy for these structures to fall into a state of disrepair until one suddenly, and often catastrophically, fails. We’ve already looked at a precarious dam situation currently playing out in California, and although engineers have that situation under control for now, other times we haven’t been so lucky. Today we’ll delve into a couple of notable catastrophic failures and how they might be avoided in future designs.

Gaining Weight While Delaying Repairs

Most of us take infrastructure for granted every day. Power lines, roads, pipelines, and everything else have a sense of permanence and banality that can’t be easily shaken. Sadly, this reality shattered for most people in Minneapolis, Minnesota in August 2007.

I-35W Mississippi River bowed gusset plates. Photo taken on 6/12/2003, four years before the collapse. Unknown author Public Domain.

A massive truss arch bridge carrying interstate highway traffic over the Mississippi River suddenly collapsed, tragically killing 13 people and injuring another 145. Rush hour traffic and an ongoing construction project had overloaded the corroded, failing bridge to the point that the center span of the bridge abruptly fell into the river, followed by the adjacent spans. The casualties could have been much higher, too, but four of the eight lanes were closed for resurfacing.

While engineers had noted major structural issues in the early 1990s, no substantial measures were ever taken to improve the bridge. In fact, changes to the bridge may have exacerbated its issues. A de-icing system was installed in the late 90s which may have increased the rate of corrosion of the steel components of the bridge, including the bearings. Additional concrete was laid on the bridge for resurfacing throughout the years, and sturdier guard rails were added as well. Additionally, at the time of collapse, a construction crew was on the bridge resurfacing the travel lanes.

However, the structural deficiencies of the bridge were no secret. The US Department of Transportation (DOT) rates bridges on a 100-point scale, and for two years before its collapse this bridge scored a 50, putting it at the bottom of a list of bridges in the US. Additionally, a plan to reinforce the bridge was cancelled in the early 2000s because it was found that drilling into the existing structure to install the reinforcing steel would actually weaken the bridge further.

After the investigation was completed, it was found that the gusset plates — used to connect beams and girders together — were the main cause of the failure. They were undersized for the original weight of the bridge, which had increased around 20% over the course of its lifetime. Gusset plates don’t just connect the structural members of bridges and buildings (including traditional wood-frame buildings), but they make the connections stronger than they would otherwise be. For this reason it is crucial that gusset plates be sized correctly and repaired or replaced if needed. Additionally, the bridge’s bearings were partially frozen that day which may have contributed to the overloading and failure of the gusset plates, but the main cause of the failure was their overload.

Since the collapse, a new bridge has opened to carry the interstate traffic over the river. The pre-cast concrete box girder bridge was completed ahead of schedule in 2008 and includes a number of modern technologies, such as an array of sensors to measure movement of the bridge. Additionally, the concrete was poured with a number of additives that will increase the lifespan of the bridge to an estimated 100 years.

Structure Flaws Plus Human Error

While the collapse of the I-35 bridge in Minnesota was tragic and arguably preventable, it’s far from the only major infrastructure failure that has happened in recent memory. Just a day’s drive south from Minneapolis is the Taum Sauk Hydroelectric Power Station, a dam that is used as a pumped storage facility. In 2005, the dam was overtopped which resulted in its complete failure. Approximately one billion gallons of water was released when the dam wall collapsed, and the 20-foot wall of water that poured from the dam destroyed everything in its path. Luckily no one was killed as a result, although one family was swept away by the flood and suffered major injuries.

The dam during repairs. The damaged section can be seen towards the top right of the structure. Additionally, the scoured terrain from the original failure is visible. Photo by KTrimble CC-BY-SA 3.0

Part of what makes this dam unique is that it wasn’t built into any of the surrounding topography like traditional dams are, aside from being built on the top of a hill. The dam makes a complete loop to impound the water that is stored there and acts as a huge battery for the power grid. During the night when power is cheap water is pumped into the reservoir, and then is allowed to flow out of the dam during times of peak demand on the grid during the day. While the cause of the dam failure in Missouri was mostly operator error (known flaws with gauging how full the reservoir was, purposefully filling the dam beyond its capacity), these bad practices combined with construction flaws in a catastrophic way.

The dam has since been repaired, using more robust roller-compacted concrete instead of earth fill. Additionally, and perhaps more importantly, a more accurate gauge system was installed and it seems that the dam’s operators are less likely to operate the plant in the same way that they were before the failure. This is in part due to a $15 million fine levied against them by the Federal Energy Regulatory Commission (the second-highest fine ever levied by FERC).

While the failures of the I-35 bridge and the Taum Sauk dam aren’t the only modern infrastructure failures we’ve seen in recent history, they are among the worst and are a reminder that flawed design coupled will poor maintenance practices and operating procedures are a recipe for literal disaster.

If there is another piece of infrastructure near you that isn’t receiving the attention it deserves, tell us about it in the comments or email the author directly at bryancockfield@hackaday.com.

64 thoughts on “Failing Infrastructure and the Lessons It Teaches

  1. I see bean counters as partially to blame. They gain power by reducing budgets. And, never underestimate engineering elite-ism. They also gain power by reducing budgets. (Many times, that reduced budget is for “Blue Collar” maintenance.
    To coin a phrase, “The Sky Is Falling”..

      1. Some “bean counters” will eat half their beans and use the energy to plant the other half and grow more (use tax money to improve conditions for people and businesses to thrive)
        While other “bean counters” hand out their beans to other farmers and pray that they’ll share the bounty come harvest time (Tax deals for large business that inevitably move on when the deal is over)
        Still more “bean counters” live on a starvation diet and clear cut the land praying that beans just appear and then beating their children because they didn’t pray hard enough for beans (effectively removing taxes on high earners while slashing programs and putting the remaining burden on small business and low and middle income groups)
        Worse is if you do the last one the rare and beautiful “501(c)” bird will come and nest on the cleared land and nobody will let you plant there anymore, even worse you may attract the invasive and obnoxiously loud bean stealing “527” bird well know for it’s constant calling.

        What’s any of this got to do with anything? I don’t even know anymore!

    1. Having spent a career in the maintenance sector, I can tell you that this is spot on. Maintenance doesn’t make any money, it only costs (from the bean counters’ perspectives). Management pushes to reduce budgets, decreasing the ability of maintenance workers to perform necessary routine maintenance. Replacing worn out equipment and infrastructure is also frowned upon, and often completely refused. This leads to costlier (and sometimes deadly) problems down the road, which are then blamed on the maintenance departments.

      1. I’ve heard a lot of times “oil is cheaper than a new engine”, but also “use it until it falls apart”, about fleets of industrial machines. I definitely think that preventive maintenance as well as regular inspections is less expensive overall and is also better for the environment.
        Most things do not fail without a warning, most of the time when overstressed (or contaminated), semiconductors’ leakage or capacitors’ ESR would increase, oil and grease would form some sludge or would have a funky color (darker than black, bubbly milkshake or glitter)…
        That’s also why I’m picky on devices that are engineered for a short lifespan (typically TVs from some brands, consumer laptops, most entry level smartphones, and parts/tools made of “Chinesium”).
        I’ve also hear of people saying they were “not rich enough to buy cheap stuff”, thinking about the total cost of ownership throughout the device’s lifespan.

        1. “I definitely think that preventive maintenance as well as regular inspections is less expensive overall”

          So does United Parcel Service (UPS). They learned the average failure time for about every component in their delivery vans and replace those components before that time arrives. For example (I might be wrong about the hours on this…) An alternator is replaced every 180 hours with a rebuilt one. They know the cost of replacing it at that time will save them costly downtime some point in the future.

          1. There’s many types of maintenance. Routinely replacing parts before average lifespan is not the most effective one, since nearly all of the parts have a lifespan beyond.

  2. Municipal budget management with federal oversight didn’t stop rust.
    It is only fair that the taxpayers sue to get 2 years of Gas tax revenue refunded — right?

  3. The emergency evacuation earlier this year of the lowlands of Butte County and surrounding areas was due to similar neglect of the Oroville Dam. Both the main and emergency spillways were overdue for repairs but the can kept getting kicked. In comes the biggest influx of runoff in a decade and suddenly the dam might wash out and inundate everyone downstream.

    1. The irony is that dams kill hundreds every year, destroy wildlife habitats, ruin river ecosystems, and cause methane emissions. An aging and neglected dam upstream of a city or a town is a disaster waiting to happen.

      https://www.americanrivers.org/2016/10/removing-dams-can-save-lives/

      But where’s all the concerned people picketing and marching against hydroelectric power? They’re all marching and picketing against nuclear power, which has an objectively better safety record. Go figure.

      1. Yeah. No. Total failure of a small nuclear plant would still be a massive disaster. A small damn might destroy a few house, or flood land, temporarily. The article also mention that those dam are either no longer in use, or located in dangerous location.

        No mention of any death anywhere in the article.

        Smart use of hydro is pretty efficient and safe. Smart use of nuclear energy is… still somewhat dumb because of the design of reactor and the nature of the fuel and it’s byproduct.

        1. Apples to oranges.

          How many “total failures” have there been for nuclear power? One. Chernobyl.

          And even Chernobyl isn’t a total loss – it’s just that nobody wants to go in there and clean it up. Likewise, if the Hoover Dam burst, who would foot the bill to rebuild all the flooded cities? Probably nobody.

  4. Shortly after the I-35 bridge collapse, another bridge further downstream was found with insufficient gussets, and the bridge was closed for most of the replacement. The officials were wary of having construction equipment and traffic on the bridge because of the lessons learned upstream.

    1. And that comment, in a nutshell, is why everything in this country is falling apart, People are perfectly happy to use the commons, but the moment we need to do something so it’s not destroyed, they proclaim that it’s ‘waste’.

    1. Unless that aqueduct was force to do more work that it was designed for during the past 2000 years “still standing” can’t really mean much. IMO apples to oranges.

    2. Pont Du Gard was also used as a bridge and the locals kept repairing it. The rest of the aqueduct became clogged up with silt and eventually broke down. The Pont Du Gard was eventually restored in the 18th century.

  5. The I-35W bridge was adequately constructed for the surface, railings, and traffic load at the time it was built. Government agencies ignoring that and loading it down with more layers of surfacing etc. while never doing the maintenance is what took it down.

    What should have happened was when buckling gusset plates were first discovered would have been carefully stripping off all the layers of surfacing then a total inspection followed by rust remediation, replacement of parts as needed with thicker metal. Then apply a new surface with the latest technology for waterproofing and ice melting so that water would run off the sides instead of getting down to the structure.

    While all that was going on, build a new bridge next to the old one so that when finished the old bridge and the new would each be used for one way traffic.

    But instead they went with “Eh, let’s do nothing but hope it won’t fall down while the cost to build a replacement keeps increasing – pushing that project ever farther into the future.”

    How much more did the lawsuits cost over what fixing the bridge would have?

      1. It was a 20% increase in static load. They had a full construction crew doing roadway resurfacing and rush hour traffic on the bridge on top of that.

        Plus, the bridge bearings had seized or partially seized, putting additional load on the gussets.

        Larger gussets probably would’ve just prolonged the neglect further and caused a larger failure.

    1. The I-35W bridge was not adequately constructed. The gussets had been undersized. The engineering description of the type of failure is overload, but the materials and weight gain should have been well below the strength of the bridge had the designers not skipped some critical analysis steps.

      From the NTSB Executive Summary:

      “the probable cause of the collapse of the I-35W bridge in Minneapolis, Minnesota, was the inadequate load capacity, due to a design error”

      https://www.ntsb.gov/investigations/AccidentReports/Pages/HAR0803.aspx

      From the report, Conclusion (3):

      “the following were neither causal nor contributory to the collapse of the I-35W bridge: corrosion damage found on the gusset plates at the L11 nodes and elsewhere, fracture of a floor truss, preexisting cracking in the bridge deck truss or approach spans, temperature effects, or shifting of the piers.”

      From the report, Conclusion (6):
      “The gusset plates at the U10 nodes, where the collapse initiated, had inadequate capacity for the expected loads on the structure, even in the original as-designed condition.”

      The NTSB found the original design firm failed to do the required calculations necessary to adequately determine the necessary strength.They also found that follow on inspection and analysis efforts skipped them, while concentrating on significant fatigue cracks.

      Simply substituting a thicker plate is not a good solution as it can result in load transfer to some other location – it’s a form of structural failure whack-a-mole. I’ve seen this in cases of fatigue failures where additional material just moves the location of the cracks, but doesn’t prevent them. A good solution starts with an examination of the original stress analysis to see where it failed to predict the current condition so that any new analysis can be checked to see that no new errors have been added.

  6. Our maintenance guys are loaned to other departments to do unskilled labor so that they “earn” money for the group. If they’re actually fixing our machines, our department loses revenue.

  7. Another overlooked infrastructure is the sewer line. Bury em and forget em until about 50 or 100 years later when it becomes clogged or causing sinkhole because the original line had developed cracks along the line. 3 houses were forever lost because of an old line in Fraser, MI last December. No one was killed fortunately but this could have been prevented if someone took the time to examine all the sewer lines regularly and make repair soon but those costs money and no one wants to spend money on any project that is otherwise invisible to public and usually have very little return.

    1. My city is plagued by 125 year old water mains exploding, usually ripping up a good long section of street in the process. It’s just not economical to replace them proactively; they’re all over hundred year old sections of cast iron pipe. You’d have to assume they’re all bad. That’s literally hundred upon hundreds of extremely expensive digging repairs, given 90% of these mains have been paved over a couple dozen times and have all sorts of other buried services running over the top of them.

    2. Ah yes, the sewer pipe fiasco.. Unintended consequences rule the infrastructure. L.A. County replaced aging sewer lines with “100 Year” concrete pipe.. Massive over engineered reinforced concrete.. Shortly after they were installed, the “Envior-Nazis” petitioned the county to stop toxic waste dumping, to “Protect The Sewers”…
      The county agrees and mostly stops battery re builders, chrome plating shops, and others from dumping toxic waste.
      Skip forward 3 years, and a trash truck falls through the pavement, into a collapsed sewer. Oh, the humanity.
      As it turns out, the toxic waste normally in the sewer was toxic enough to kill off most of the bacteria also there.
      No more toxic waste, bacteria “Blooms” and eats the concrete. Trash trucks fall into sewer. Now the county is really effed. Those 100 year sewer pipes are essentially “Junk” due to cleaning up the environment.. EFF!

        1. Yes, that was the “Bean Counter” answer, line them with plastic. Just the same as “Kicking The Can Down The Block”, that was simply a stop gap measure to reduce the yearly budget with unfunded emergencies.
          And just a cheap “Band Aid” on a cancerous lesion.

          1. Theatrical language aside as long as the structural integrity of the pipe is still there, lining is perfectly acceptable, and has been for several decades.

  8. My home state of Louisiana is currently dealing with this very issue. The state is over a billion dollars in debt, and so road and bridge maintenance has been suffering more and more. The Huey P. Long bridge, one of the few bridges that allow trains to cross the Mississippi River into Baton Rouge (the capital), has visible fissures and cracks in the support pylons, and yet it’s an uphill battle to get the money for constructing a new bridge.
    God forbid you mention raising taxes to foot the bill, you’d have an angry mob on your hands.

    1. “God forbid you mention raising taxes to foot the bill, you’d have an angry mob on your hands.”

      They are already taking enough of our money in taxes. They are simply spending it in the wrong places, such as their pet social programs.

        1. It’s not the federal government’s job to feed the poor. Quit forcing your religious beliefs on the rest of us and taking away the fruits of our labor with the government gun to our heads and giving it to someone else.

  9. Additional loads on bridges can cause all sorts of un-forseen issues, http://www.heraldscotland.com/news/12052777.Kingston_Bridge_set_for_a__14m_lift_off/
    This work was carried out on a major structure in Glasgow (Scotland), one of the busiest road bridges in Europe and led to quite a few quips, such as always remember to pack your life jacket before crossing the Kingston bridge.. fortunately it went more or less without a hitch, although the project did come in somewhat over budget, but nowhere near the cost of replacement.

  10. Sadly this problem is not unique to USA. For example several European capitals have nasty record of gas leaks blowing up buildings or just generally polluting. Railroads and highways are mostly ok, but our power grid is severely stretched thanks to surges from renewables.
    We often prevent blackouts just at the last moment.

  11. Why temp people to use up their energy whining about it here… here where it will do NO GOOD? Post some addresses to write to their local issues… and give us a list of the issues local to us… and how to discover more ourselves. Pokin’ at the ppl does no good when what should be happening is the ppl given info or how to find it themselves to poke at the responsible leaders. Anything else is just noise.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s