5 of History's Worst Engineering Disasters

While collecting this short list of engineering disasters, I decided that to qualify as a disaster, an event had to involve loss of life, and the larger the loss of life, the larger the disaster. This ruled out a couple of famous engineering snafus, such as the “swinging” Tacoma Narrows Bridge that suffered a wind-induced collapse in 1940 (as seen in this famous video). It also disqualified NASA’s loss of the $125-million Mars Climate Orbiter in 1999 that made it close to Mars before an engineering mistake in not converting pounds of force to newtons doomed it.

An engineering disaster also had to hinge on engineering. This ruled out catastrophes such as the 9/11 Twin Towers collapse caused by a terrorist attack and the building of the Panama Canal in which an estimated 30,000 workers were killed, most by disease.

So here’s my list (in reverse chronological order):

The Chernobyl Nuclear Plant Explosion: Structurally unsound reactors failed, leading to a string of explosions and widespread radioactive fallout in 1986. Up to 64 people died during the event, and the resulting escaped radiation led to another 4,000 deaths and 30,000 more people suffering premature cancer deaths.

The Collapse of the Hyatt Regency Hotel Walkway: A change in the original design of a series of walkways suspended above the lobby’s hotel led to walkways falling into the crowded lobby. 114 were killed and another 200 injured. Engineers who had approved the final drawings were convicted by the Missouri Board of Architects, Professional Engineers and Land Surveyors and lost their engineering licenses.

The Banqiao Reservoir Dam Collapse: In 1975, a once-in-a-millennium flood came roaring down the River Ru in China. Engineers had not planned for such an event and the dam failed. Over 700 million cubic meters were released in six hours, killing at least 171,000 people and leaving 11 million people homeless.

The St. Francis Dam Disaster: To meet the needs of a rapidly expanding Los Angeles in the mid-1920s, the city hired William Mulholland to design and build a dam along the L.A. Aqueduct. Mulholland did the design work himself and oversaw construction. For two years after the dam was completed, he is said to have ignored warnings that the dam had developed cracks and was leaking. In 1928, just two hours after Mulholland himself inspected the dam, it burst, killing 450 people.

The explosion of the steamboat Sultana: In 1865, the captain of the passenger steamship Sultan ordered quick repairs to a leaking boiler. The person who did the repairs told the captain that the boiler shouldn’t be considered safe and it needed more inspections. But the ship’s chief engineer approved the repairs and the ship continued on the Mississippi River. Two of the ship’s three boilers soon exploded, setting the wooden boat on fire. 1,547 people lost their lives, more than were killed in the sinking of the Titanic.

What do you consider the world’s worst engineering disaster? Leave a comment or send us an email.

Discuss this Blog Entry 14

on Sep 16, 2015

What is your criteria for awfulness of disaster? I would think that the Banqiao Reservoir dam collapse should be at the top of your list. How about Bhopal? Storing Methyl Isocyanate must be stored in refrigerated tanks. This was a very bad idea in India.

on Sep 18, 2015

Ironically, Union Carbide had declined to stay on in 1984 as operating contractor for the nuclear facilities in Oak Ridge TN. Their legal department felt that there was too much associated liability. This was just a few months before Bhopal.
For years thereafter Carbide officials were afraid to even travel near India for fear of being arrested.

on Sep 30, 2015

OK, now I'll jump in with my two cents: I have worked for the last 36 years in engineering, keeping a strong personal interest in safety and accident avoidance by careful design practices. Years ago (1984), desiring to raise my somewhat constrained salary, I visited Unión Carbide offices in my country, as a good friend who worked for them invited me to get an interview with their engineering bosses, because they were hiring engineers and offered to triple my salary if I met their requirements. I solved the short exam, demostrated my command of english language and fully met the profile. BUT, as I was finishing the fourth or fifth meeting with them, I asked their top engineering manager about the Safety criteria and policy of them... To my big surprise he told me, with a large dose of effrontery: "look, whatever it happens, it NEEDS to be kept INSIDE the perimetral wall of our installations, first; and second; not to be broadcasted in the news... that's all my friend". What He was wanting to tell me, was that all my experience and special interest in a serious culture of industrial safety and safety engineering was NOT particularly appreciated, wanted or asked by UC from me. In other words, "We are not phanatics of safety, what we want from you is to lower costs and keep production running, that's it"... this conversation happened just a couple of weeks before their catastrophic accident at Bophal. This revelation made me lose a lot of the enchantment and inspiration I had in becoming an employe of UC, regardless of their salary offer. Fortunaltely, as they kept delaying their bureaucratic hiring process, I decided to walk away and declined any further offer from them, and consider to this day that I was really FORTUNATE of NOT becoming an employee of such a reckless and human-life disregarding company, and feel much better for my decisión NOT to work for them! As for my Friend that worked years for them, he was terminated not one, but THREE times during the next 15 years, only to be re-hired but with lower and lower income. So much for UC or PRAXAIR. Just take a quick look at www dot endgame.org/carbide-history.html to get an idea of their safety record. and you are right, in march 1984 they ceased operating the U.S. Department of Energy's Oak Ridge, Tennessee (begun in 1943) and Paducah, Kentucky facilities, and dissolved its Nuclear División.

on Sep 25, 2015

good,tôi cũng nghĩ thế

on Sep 16, 2015

I gave my criteria for an engineering disaster and noted that I was listing them in reverse chronological order, not in order of awfulness. Thanks for chiming in with another disaster, the Union Carbide/Bhopal pesticide plant leak.

on Sep 21, 2015

The crash of Turkish Airlines Flight 981, a DC10, on March 3, 1974 caused the death of 333 passengers and 12 crew members. The design of the cargo door latch prevented those operating the latch from verifying that it was properly secured. The unsecured cargo door blew out resulting the collapse of the main cabin floor disabling the pilots' control of the plane. Tragically, the flaw had been detected during test flights, but not corrected due to the potential cost of the repair.

on Sep 21, 2015

Engineering disasters that are NOT foreseen is one category.

Engineering disasters that are foreseen and ignored for selfish reasons are most damning.

Management interference usually causes the later.

Cost-Benefit Analysis

Based on the Crash tests that had been done, Ford knew the Pinto posed a serious fire hazard. One of Ford's former engineers testified that the highest levels of management at Ford understood the danger but decided against making changes due to the potential costs.

There are legal grounds for performing a cost-benefit analysis. In the 1947 case of United States v. Carroll Towing Co., Judge Hand effectively devised the cost-benefit analysis to help decide future cases. (Leggett) Additionally, the NHTSA, a federal agency, approved the use of a cost-benefit analysis for determining car design safety standards. For the purpose of its cost-benefit analysis, Ford determined the value of human life at $200,000 per fatality. Thus, the decision to issue a recall and install the necessary safety measures was weighed mathematically against the benefit of preventing a certain projected number of deaths.

Ford projected a cost of $137 million to fix the problem with only a $49.5 million benefit (notice that Ford's projection was based on the prediction of 2,100 burned vehicles and 180 burn deaths). Ford determined that it would be much more costly to fix the problem in its entirety than to pay for the few cases that resulted from its poor design. “The Ford Motor Company's risk/benefit analysis indicated costs would be 2.5 times larger than the resulting benefits” (Legget). Ultimately, Ford decided not to implement the fixes in the development stages and continued to forgo the fixes once the car was on the road. (Birsch)


Harley Copp was a former Ford engineer who has testified that the highest levels of management at Ford understood the Pinto's problem with the fuel tank in crash tests but were unwilling to deviate from the design if the change would cost money.

Victim Action

In September of 1978 Ford Motor Company was indicted by an Indiana Grand Jury for three felony counts of reckless homicide. The case resulted from a rear-end collision between a van and a Ford Pinto. In the car were three girls, who all burned to death. People on the scene claimed that the collision occurred at a relatively low speed. The significance of the Indiana trial was that it was the first time a corporation had been charged with criminal violations rather than simply civil violations.

When the trial concluded in March of 1980 Ford was judged to be innocent of all charges. The last Pinto was produced in 1980. (Birsch p.5-6)

fordpinto. blogspot. com/2010_03_01_archive. html

on Sep 22, 2015


Your summary of the Pinto disaster brings back the memory of one of my professor's using the Pinto gas tank as an example of innovative cost reduction. I think that this was in 1970. How wrong he was!

Roger Weise

on Sep 30, 2015

Engineering may fail in some situation. But, we know that the technology will go utmost, but have limitations as well. Nothing can be perfectly done, the manufacturers leave drawbacks for further advancement of technology. These disasters might be of that type, but these can be overcome today. However, the mistake we make will get solved in next decade.

on Sep 30, 2015

"Nothing can be perfectly done"...

What a perfect (pun intended) pretext to do things carelessly. Most of the engineering disasters have ocurred because of what is called "complacency": a feeling of smug or uncritical satisfaction with oneself or one's achievements... either during design or operation of the installation.

True; you cannot design totally "Fool-Proof" as there will be always be an even smarter fool that will detonate the accident. But my point is that in order to truly design a safe installation, you need to get a full safety culture across all the design team. (and avoid the ever-present "bean-counters" from making their snafus).

on Oct 1, 2015

Thanks yoy very much

on Oct 1, 2015

Very important theme, Stephen. I would include at least three Offshore platform disasters to your list: First the Piper-A disaster in the English sector of the North Sea back in 1988 with 165 out of 266 persons on board dead, that has been heavily studied (and it is astonishing that nowadays some recommendations given in the extensive "Lord Cullen" report of the accident are still NOT applied to many other newer platforms around the world!).
Next I would mention the Deepwater Horizon accident in Gulf of Mexico waters in 2010 (a BP and others disaster), which only produced 11 deaths, but the worst environmental catastrophe of all the oil spills in history.
One thing that is completely overlooked in the many reports and articles about that disaster, is that none of them clearly identify what I call "the point-of-no-return" in the accident chain of events: the second and largest explosión that doomed the platform, caused by what I consider as a grave engineering omission: the diesel engined electrical generators aspirated natural gas from the leaking well, thus getting their fuel from the air, which caused them to inevitably (given the platform design) overspeed and self-destruct, causing the largest explosión that ultimately led to the platform sink into the ocean. Would its designers had provided air inlet cutoff valves actuated by the combustible gas detectors, those engines would have prevented their overspeed and explosive mechanical breakage. Accidents are the very best learning opportunities for improving design and engineering, but many times this opportunities are ignored or worse: occulted from public view. Another platform accident that was poorly reviewed happened in Mexico´s Campeche Bay, when an old Jack-up type platform (called "Usumacinta") lost stability during a strong storm and struck a smaller fixed platform, breaking the wellhead "Christmas Tree" and causing a lethal sour gas leak. In the following emergency abandonment, several of the (supossedly safe) life salvaging boats either sunk or capsized or were torn to pieces, drowning an undisclosed number or people. As in many other cases, the official reports fail to find the true root-cause (in this case most probably was the type of base plate of the platform: a large flat surface that caused hydrodynamic scouring of the sea bottom when the platform started to rock side to side, which allowed the whole platform to reach more and more inclination, until its drilling cantilever knocked-of part of the flowing well Christmas Tree on the fixed platform, unleashing the gas leak that obligued the oil workers to abandon the large Jack-up platform in a terrible sea state conditions. Another thing that seems forgotten in this acident, is the terrible damage that the sea inflicted to supossedly "safe" life-saving, fire-proof, covered life boats, which were either poorly designed, built or maintained, probably because of being of resin and fiberglass construction and subjected to years of tropical sunlight degradation became too fragile.
Outside of offshore installations, interesting and grave accidents that can be on the list are: 1) The BP Refinery at Texas City in 2005, with at least 15 fatalities and 180 injuries, at the time it was the largest BP refinery and the third largest in the USA, and interesting because is was exemplary in the worst sense, having registered at least 23 fatalities over the previous 30 years, including three in 2004 alone. This wasn't even the first time that this specific safety compliance lapse had occurred with this very equipment. There had been eight serious releases of flammable material out of that stack in the years prior to 2005, but nobody had ever investigated the causes. Moreover, such overfilling incidents in distillation columns had happened in other refineries, and investigations cited in industry literature stressed the importance of accurate level measurements as part of overall safety risk management programs.
2) The two large Nuclear Power Generating plant accidents, at Chernobyl and Fukushima of course! (I don't want to stir hot debates over the -supossedly- extremely "safe" and "fail-safe" designs, nuclear engineering frequenlty touted as the best example of utmost "safe" designs, but if the petroleum industry, even with its large resources still produces accidents that have some component of less-than-stellar engineering, Nuclear is not too different, no offense intended. Amclaussen.

on Apr 9, 2016

This has to be at the top of the list

How is burning wood pellets made from US forest trees, shipped 4,000 miles, to power the UK National Grid, a solution to climate change?

Burning wood releases more CO2 than burning coal - This is a case of the solution being many times worse than the problem ...

How can deforestation + shipping millions of tons of wood per year + burning wood be a solution to climate change?

Enviva, Zilkha Biomass, Highland Pellets, UK Drax, UK National Grid and many others are participating on a sham, a false solution to the most urgent problem in history

on Jun 9, 2016

burning wood produces LESS CO2 than burning coal.
It's down to the longer/shorter CH chains, changing the proportions of Carbon to Hydrogen.

Please or Register to post comments.

What's A Skeptical Engineer?

A questioning, sometimes humorous look at technologies, engineering, and the world.


Stephen Mraz

Steve serves as Senior Editor of Machine Design.  He has 23 years of service and has a B.S. Biomedical Engineering from CWRU. Steve was a E-2C Hawkeye Naval Flight Officer in the U.S. Navy. He...
Blog Archive
Connect With Us

Sponsored Introduction Continue on to (or wait seconds) ×