What should be in a reading list on failures, disasters and their prevention 12

[MartinLe]

Beyond the specific codes etc. I work with in my field, what are essential readings on the above?

I'm interested in the social aspects as well as in how to systematically think about the technical aspects. Reports of investigations into specific accidents may also be interesting.

My own field is supervision of construction sites and design of wastewater treatment plants but I'm also interested in safety thinking for other fields.

Onn one hand, I sometimes encounter codes or standards that don't feel entirely logical, but then I'm not sure what would actually be logical. On the other hand, there's often a drive to not follow all regulations etc. when they get into the way of the work.

What would I do on such a reading list?
I'm thinking of getting "Normal Accidents" unless a more recent work on similar themes is better in some way.

A few years back I read a longish report on the sinking of the Merchant Vessel Faro that went into the technical details but also contained lots of transcripts of what was said on the bridge, alowing one at least to guess what went through the minds of the sailors in their last hours (don'tfind the link to the report now, shorter article without technical details:

https://www.vanityfair.com/news/2018/04/inside-el-faro-the-worst-us-maritime-disaster-in-decades)

This subforum!

 

[thebard3]

Speaking of space shuttles, the Columbia Accident Investigation Board report is voluminous and extremely interesting. Lots of parallels to the Challenger accident but different in it's own way. As with a lot of accidents, hubris is a significant contributor.

https://www.nasa.gov/columbia/home/CAIB_Vol1.html

Brad Waybright

It's all okay as long as it's okay.

 

[Eufalconimorph]

How Complex Systems Fail. 5 pages, VERY worth keeping in mind.

 

[NorthCivil]

The challenger disaster, with the O-ring issue, the management rush to meet the launch date, despite engineers knowing there was an issue without a solution implemented, was frequently raised throughout my university education in ethics, law, and management courses.

 

[Bill West]

Hi MartinLe. Your view point is very interesting and similar to my own safety concerns.

I take it that in "codes" you are sort of referring to the technical prudence that is part of practicing professional engineering. And in "social" you are sort of referring to the construction site attitudes that are part of being a "professional" journeyman. For the former I think the others have made good suggestions. For the latter two sources have influenced me, both are from my nonprofessional interest in railroads.

-The Historical Interstate Commerce Commission reports available online have perhaps a thousand railroad accident investigations over more than half a century. A very persistent thread in them is how many times two to four minor oversights of the rules added up to a serious accident. Gross violations were far from being the most common factor.

-TrainOrders.com is a hobbyist/staff/executive railroad discussion board. Although their "Nostalgia" forum threads that hit me are much dispersed there, the essence seemed to be the staff jesting about past cutting of petty corners and the executives retrospectively criticizing them strongly for it. In the end I felt that the staff got away safely 99% of the time and so thought it was okay but the executives saw the 1%'s cumulative loss in dollars, sick time, permanent disability and most sadly, death. The difficulty seemed to be in convincing the staff that any cutting of corners would eventually backfire and the executives just didn't want to be part of that.

Bill

 

[MortenA]

Piper Alpha -Theres a pretty good movie made - but this should also be OK:

https://www.youtube.com/watch?v=Nwbw5PHZnqk

 

[Trenno]

Great podcast by Sean Brady on all kinds of failures and disasters. He explores how human factors are critical in most of these events.

Sean Brady Podcast

 

[MartinLe]

How complex systems fail is IMO a good read and it's interesting to go through the bullet points and see if and how they are baked into different codes and design processes. And what it would mean if we take "There's no root cause" seriously.

So far no one commented on "Normal Accidents" too sociological and so no one here read it?

Bent Flyvbjerg has lots of papers online - is the Megaprojects book a good starting point, or rather some of the papers?

 

[LittleInch]

One thing I'm trying to do is refer to anything that happens as an "Incident".

As soon as you use the term "accident" it implies that the cause was "accidental" and hence somehow not in our control or ability to influence things. It might only be a subtle difference, but in the same way I ban the use of the word "Temporary" from any description of work.

The inherent implication is that a "temporary" design or item or connection doesn't have the same level of design, construction, quality and scrutiny as the "permanent" design.

If it contains pressurised fluids or bears a load then it doesn't matter if it's in operation for one second or one decade. It can still kill you.

Thus are subtle changes in how people think about risks and incidents.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ACtrafficengr]

A lot of "user error" is actually caused by poor design. For a decent primer on human factors in design, I suggest The Design of Everyday Things.

My glass has a v/c ratio of 0.5

Maybe the tyranny of Murphy is the penalty for hubris. -

http://xkcd.com/319/

 

[JohnRBaker]

If we're going to talk about 'product design', then I would suggest watching the documentary, 'Objectified', produced by Gary Hustwit:

https://www.imdb.com/title/tt1241325/

And while you're at it, despite them not fitting this thread, try watching Hustwit's other documentaries, including 'Helvetica' and 'Urbanized':

https://www.imdb.com/title/tt0847817/

https://www.imdb.com/title/tt1701976/

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

 

[msquared48]

The Hyatt Regency Walkway failure.

Mike McCann, PE, SE (WA, HI)

 

[VEBill]

The book 'Fatal Defect' by Ivars Peterson explores buggy software.

 

[CWEng]

As for other fields and safe thinking, the Nuclear industry has a lot on human performance, which could apply to operating any plant, including WWTP.

On example would be:

http://nuclearsafety.info/wp-content/uploads/2010/07/Human-Performance-Reference-Manual.pdf

 

[VEBill]

This very forum.

 

[ATSE]

OSHA
OSHA produces 20-50 page documents with site photos of all construction accidents.
The target audience is semi technical.
Example: "Investigation of the December 6, 2017 Fatal Parking Garage Collapse at Berkman Plaza 2, Jacksonville, FL"

Unlike many politically correct government documents, these documents identify negligence and the guilty parties by name (usually the contractor and engineer share culpability).

 

[RolMec]

NASA does safety messages on various incidents, always with an edge towards everyday's engineering works.
This one is special to me:
Link
as it conveys, how human decisions influence the course of acion.
I remember that time, the incredulous facts that the maintenance workers didn't heed the checking systems red light for that wheel, .. "because the reading could not possibly be right" and that the passenges and staff did not pull the emergency brake even with a heavy metal spike having pierced the compartment floor.



Roland Heilmann

 

[Bill West]

Sorry Roland, but I find the wording of that memo a bit weak.

3 km at 200 kmh leaves about a minute for the passengers to instinctively get out of the way, realize the situation, go back 2 cars to find the manager, tell him, let him get back to the scene to gauge the info and call the driver. This unfortunately falls on the "insufficient time" side of the analysis rather than anyboy's inaction.

Yes the "heritage" wheel was designed about 1935 for an extremely successful, 40mph 17 ton car. But heritage is not the word that is wrong in the thought, a successful 5 year old 40mph 17 ton design would have also been a question. It's the checking of the application that should have been the emphasized phrase.

There are varying degrees of completeness and objectivity in accident reports. Three drivers in an engine cab missed a signal in Ontario a few years back. They were lined for a different track than they had expected and were killed in the ensuing accident. The official report goes over all sorts of issues but nowhere does it come out and say the fellows just sadly missed the signal.

Bill

 

[IRstuff]

The link is really a lessons learned briefing from 9 years after the accident, and is structured like an elevator briefing.

https://en.wikipedia.org/wiki/Eschede_derailment

suggests that the root cause was anticipated well before the accident, in 1992, but no one did anything about it.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[RolMec]

Yes, there's much more detailed investigation to be found.
Actual distance from rim-pierces-compartment event to derailment was ~6 km, almost 2 minutes time (wiki)
And yes, the manager was obliged by regulations to first assess the situation.
However, what would one decide to do if a really big bang makes bulge up the floor in a train compartment?
However, how would one decide if the readings on a train wheel checking machine could not possible be real?
I find technical and social aspects intertwined, "internal" logical argument going against conformity, belief in regulation, pressure felt from schedule...

Roland Heilmann

 

[pmover]

among the other listings, the book "What Went Wrong" by Trevor Klatz.

https://www.elsevier.com/books/what-went-wrong/kletz/978-1-85617-531-9

i read an edition back in the early 90's. a worthy read.