Engineering disasters are always the result of bad mgmt.... 5

[leanne]

Engineering disasters are always the result of bad management and never the result of bad engineering—or almost always. Norman F. Simenson of the Federal Aviation Administration

http://www.stsc.hill.af.mil/crosstalk/1998/04/disasters.asp

Engineering failure mechanism is defined by money, performance, physical, process, people

It can be argued that an engineering failure is:
[ol]
[li] A design that does not work (no design solution with the right performance, cost, time, specs...): [ul] [li]fusion reactor, [li]interstellar flight, [li]electric car [/ul][li] A design that works but is undesirable: [ul] [li] Beta VCR, [li] Edsel, [li] DEC PCs; [li] Aswan Dam (effect on the ecology of the Nile Basin makes it a failure)[/ul] [li] Ooops: [ul] [li] Boston's Hancock Building (glass windows blowing out); [li] Tacoma Narrows Bridge (collapsed due to wind-induced vibrations, 0 human deaths); [li] Challenger (7 dead), [li] Columbia (7 dead), [li] Apollo 1 (3 dead); [li]Apollo 13 (no deaths, overheating of Oxygen Tank No. 2 in service module external bay) [li] Ford Pinto (number of deaths unknown, caused by an $11 gasket that Ford mgmt decided not to use because of costs), [li] GM Trucks (unknown number of deaths), [li] Firestone tires on Ford SUVs (unknown number of deaths), [li] Hindenberg (35 deaths, mostly people jumping); [li] Bhopal, India (~4000 dead); [li] Walnut Street Bridge (Harrisburg PA, 2 spans washed out); [li] Quebec Bridge (2 collapses during construction, number of deaths not found in search); [li] Arroyo Pasajero bridge on I-5 (7 deaths); [li] Texas A&M bonfire (12 deaths); [li] Kansas City Hyatt (114 dead); [li] Denver Airport Baggage System (0 deaths, design issues that caused serious budget over-runs, not to mention bad press for design world); [li] California Mud Slides (lost housing, unknown numbers dead); [li] Chernobyl (62 reported deaths, 18 mile radius evacuated) [li] 3 Mile Island (near meltdown, 0 deaths per NRC website); [li] Estonia Feery (800+ deaths, shipbuilder underestimated how strong the bow visor lock should be); [li] Hartford Civic Center Coliseum (no known deaths, collapse of space-frame roof during snow storm)[li] Banqiao and Shimantan Dams (official estimate of 26K deaths, chain-reaction failure of dams in typhoon) [li] Tay Bridge (75 deaths, caused by high winds); [li] Hubble Space Telescope (0 deaths; spherical aberration error); [li]Mars Lander (0 deaths, unit lost in space due to a math error - danger Will Robinson, use a single measurement system) [li] Titanic ( 1,513 deaths; hull divided into 16 watertight compartments. Design allowed for four being flooded without endangering the liner's buoyancy - it was considered unsinkable. Compartments were not sealed off at the top, so water could fill each compartment, tilting the ship, and then continue spilling over into the next one)[/ul][/ol]
From

http://www.matscieng.sunysb.edu/disaster

Primary causes of engineering disasters are usually considered to be (I) human factors (including both 'ethical' failure and accidents), (ii) design flaws (many of which are also the result of unethical practices),
(iii) materials failures, (iv) extreme conditions or environments, and, most commonly and importantly, (v) combinations of these reasons

Comments? (other than: leanne, you obviously missed many engineering failures which my (polite) response would be, yes, but these are particularly well-recognized examples)

http://www.icivilengineer.com/Failure_Watch/

http://www.invitingdisaster.com/pages/540454/

Hopefully, we've learned from each disaster so the mistakes are not repeated. A boiler explosion in Brockton Shoe Factory in Massachusetts in 1905 leveled the factory and resulted in 58 deaths and 117 injuries - as a direct result, 10 years later ASME's Boiler and Pressure Vessel Code was published.

 

[IRstuff]

Actually, I'm convinced that it's in a tight race to get worse.

While the tools and models are better, there are fewer engineers involved, because of "productivity gains" and less relative money to begin with.

While I'm not one for NASA bashing, their recent flirtation with "faster,better" shows that while high quality work can still be accomplished, the opportunity to make mistakes also increases.

Many of the newer projects are substantially more complex than projects from 10 years ago, but with fewer cognizant engineers and a lowered amount of retained tribal knowledge. The complexity alone would be challenging, but the decreased resources in money and bodies means that less analysis and oversight is performed.

TTFN

 

[HamishMcTavish]

I'm afraid that I agree with IRStuff, complex systems (and their interactions) are a breeding ground for unrevealed commissioning faults and latent faults caused by poor operational control of plant.

A side point is that when (not if!) other disasters do happen the press tend to make a hash of reporting it. There is usually lots of coverage (i.e. hypothetical wonderings about how it happened, what the implications are etc.) but when the ACTUAL causes are revealed the press have moved onto more 'interesting' topics

HM

No more things should be presumed to exist than are absolutely necessary - William of Occam

 

[TheTick]

Don't leave it to the press to reveal the truth.

Deinition of journalist: a person who spent four years of college learning to write at an eighth grade level.

Good and evil: wrap them up and disguise it as people.

 

[PSE]

leanne,

I would agree more with SUNY Stony Brooks definition of causes rather than Simenson's.

I do not think that engineering disasters can be eliminated but I think we are doing a much better job at minimizing their potential (at least if 3 gorges holds together ). New technologies and scientific advances will also lay groundwork for new opportunities for failure (and success). The true disasters are those we do not learn from and allow to repeat.

Regards,

 

[plasgears]

Engineers need to regain the lead in engineering design and management. There are monsters out there threatening the public good by employing unqualified individuals to perform engineering work "at economical cost." It is false economy to employ promoted draftsmen to fill engineering positions. This practice is prevalent in the automotive sector.

It is a threat to the company's future, the future of the industry in question, and a threat to the nation. The uninformed public trusts that products have been designed properly by people with the "right stuff." When they find out otherwise by means of legal action, then things will change.

 

[vpl]

Plasgear - I question some of your wording. I have met some "promoted draftsmen" who quite honestly had more brains and skills than some degreed engineers that I know! Having a degree doesn't grant you common sense (i.e., I can calculate this to the 25th decimal point, so I know it's good!) Not having a degree but with lots of experience isn't necessarily great either - there are reasons to go learn all that theory. I think the message that we need to hear is "don't get complacent; don't cut corners."

Leann - I disagree with Simenson's statement on general principles: "always" is too broad. I don't like absolutes. For example, the Mars Lander was definitely an engineering mistake; while the Challenger was a management one. Others, it's harder to draw the line.

As for learning from our mistakes? Well, as George Santayana said, "Those who cannot remember the past are condemned to repeat it."

Patricia Lougheed

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[leanne]

Patricia, for most rules, there is an exception, and invariably, I get slapped with the exception after I've made a generalization. It took me a while, but I've learned to make it a point to avoid sweeping generalizations using the words ALWAYS, ALL, EVERY, or NEVER (notice I avoided saying "for ALL rules" above ) and phrases like ALL THE TIME.

Not only should we not get complacent or cut corners. We should continue to ask, is this the best way using today's currently available technology to do this in a cost-effective manner meeting our customer's requirements?

George Santayana's quote is a fitting follow-up to the last paragraph in my opening post to this thread.

 

[IRstuff]

I'll pipe in with the claim that both Mars Lander and Challenger were both failures of both engineering and management.

In the case of the Mars Lander, there were numerous opportunities for engineering to have discovered the interface error before and during the flight. Management failed to push for full design verification, possibly due to the low-cost approach.

In the case of the Challenger, design requirements were clearly violated. Engineering used tools that were never designed for task of simulating impacts by the insulation nor were the tools ever validated. Management allowed themselves to get lulled into complacency based on nothing bad having happened prior to Challenger.

TTFN

 

[RDK]

Is there really a difference between bad management and bad engineering?

Here in Manitoba our engineering act defines the practice of engineering as the skilled application of science to solving problems. It goes on to specifically include the management of engineering activities as an engineering activity on its own.

Therefore the decision to relax checking and verification is not specifically a management decision, it is a decision of engineering management and as such an engineering decision.

To discuss if it’s a management or engineering decision is simply a waste of time because it is all by definition, at least in Manitoba, the practice of engineering.

It may be easy and safe for the technical engineers to blame the pointy haired managers, but at some level the managerial types are practicing engineering as well. If the technical types have a problem with management decisions then they have an ethical obligation to become whistle blowers and bring adequate scrutiny to the issue.





Rick Kitson MBA P.Eng

Construction Project Management
From conception to completion

http://www.kitsonengineering.com

 

[bvi]

Since management is ultimately responsible (including, for the decisions as to which engineers to hire to do the work), one can take the position that engineering disastors are always the result of bad management. Management establishes hiring practices, QC procedures, continuing training, etc. etc. You know the phrase, the buck stops here. I suppose if a sole engineering consultant practioner made the mistake, you could say it was engineering, but then again, he or she was also management. Or it could be management of the firm that hired him.

 

[IFRs]

Risk is a necessary component of progress, and I believe in progress.

Risk being there means that sometimes we win, sometimes we lose and sometimes we squeak it out.

Our job as engineers and/or managers is to make decisions that accept risk but minimize the downside.

If we wait until all our ducks are in line then nothing will get done. Look at NASA, the Wright brothers, Sikorsky, Galileo. I'm proud of all of them.

I'm especially proud of NASA, for despite the failures there have been unequalled successes, astonishing experiences and boundless human growth both intellectually and spiritually.

Our jobs as engineers is to encourage forward thinking by management and other engineers, to step forward and make the future brighter. Mistakes get made. Learn from them and move on.

If we think that management are the problem, then let's become managers!! Place blame where it belongs if you can but there are lots of grey areas and I don't think its worth the effort to draw lines in the sand.

Every engineer is a manger of sorts but few managers can claim an engineering background.

So much for my 2 cents...

 

[TheBlacksmith]

I finally have to weigh in and agree. Down my way in Tidewater Virginia area, a new section of Interstate 64 may have to be torn up and redone as it does not have (any) pitch to drain water off the driving surface. Concensus is that the plans didn't provide for any pitch, the contractor dutifully brought it to the attention of the Virgina DOT (VDOT) field engineer, who forwarded to headquarters in Richmond VA (Management central), where it sat since last November. Meantime, the contractor steadfastly poured all the concrete, having no guidance to the contrary and not wishing to ensure late penalties. Well now the head of VDOT has resigned/been fired and we have an unusable roadand a few more months of congestion. I can't help but feel that the field engineer and contractor couldn't have fixed this, were it not for management. BTW the new section is high occupancy vehicle (HOV) lanes, another management fix foisted upon an unwilling populace. (I carpool with two other guys as much as our schedules allow, but 90% of the peopel in this area don't.)

Balcksmith

 

[funnelguy]

I was working on a refurbish/retrofit for a piece of equipment used in the assembly of toaster ovens. The fixture plate on the top of the newly fitted CNC indexer was about 6 feet in diameter. An OEM technician for the indexer was pacing the unit so as to apply his blessing to it when it suddenly just “ran away”, running up to the max continuous RPMS the drive could manage. I happened to be under the unit when this happened (yes, I know it was stupid) and laid motionless waiting for the unit to coast down after the technician hit the E-stop. Rivet anvils flying everywhere. I asked the technician to explain what happened and how he would prevent the same thing from happening in service. His later explanation to me was that the LED on the resolver had failed. The drive was evidently set up to count light pulses as the method of determining speed/position. LED burns out and the controller sees speed as zero. Applies max V and I to get things moving again…

How is this a manager’s failure? Is it because he/she hired someone who was so inept that they didn’t consider failure modes? I would suggest that the manager failed only if this was a repeat scenario which had not been handled properly and/or if the bozo responsible for that control system was not fired or at least retrained.

 

[PSE]

Well the accident investigation board for the Columbia breakup is not favorable to mangagement.

thread769-68850

 

[Sparweb]

A quote that comes to mind goes:

Engineering is the art of molding materials that we don't completely understand into shapes we can't precisely measure to withstand loads we can't directly assess, in such a way that it's perfectly safe for the general public.

Read enough aircraft accident accident investigations, and you quickly get the feeling that it takes many simultaneous problems or failures to bring transport aircraft down. Now that you have that impression, keep reading more, and you then begin to notice that the investigators don't always know what brought the aircraft down, and start grasping at straws.
Sometimes, the failure is in the design, sometimes the manner of operation, sometimes maintenance, sometimes the pilot just did something stupid. When an exact cause is unknown, pilot error is too often the fall-back position.

Design flaws do bring aircraft down, but I've never heard them linked to unethical practices, as the guy from Stony Brook suggests. The problem is with the "incomplete" knowledge that designers must work with as they squeeze the margins of safety tighter and tighter to save weight.

Flying is and always will be about risk management. Good points about risk were made by IFR's on this subject, unfortunately there do not seem to be any means to clearly communicate risk to the general public without causing a frenzy. Things are treated as either perfectly safe or junk.

The CAIB seems to be burning everyone up and down at NASA. Fault is found not just with management, but with weak kneed, myopic inspectors, and against congressmen (with roughly the same qualities).


STF

 

[leanne]

Building on what sparweb wrote:

Like tolerance stack-ups: missed inspection points, management decisions, under-designed, over-designed, operator error, lowest bidder not understanding specs, material failure due to fatigue - any one of these alone may not be sufficient to cause a system failure, but when they are combined...the equation can & often does change....

Failure analysis is rarely easy.

 

[patdaly]

funnelguy writes:
"How is this a manager’s failure? Is it because he/she hired someone who was so inept that they didn’t consider failure modes? I would suggest that the manager failed only if this was a repeat scenario which had not been handled properly and/or if the bozo responsible for that control system was not fired or at least retrained."

I would also go along with you. The only problem is there are entirely too many Managers who do not have the technical knowledge to manage effectively.

My horror story, I was interviewing for a Design Engineering job with a local sewer sucker manufacturer. As things progressed, they began to show me some of their designs, one of which was a 3 foot Aluminum centrifugal fan,spinning at up to 4700 RPM, which had the fan blades riveted on. I immediately stepped back and asked who had run the calculations on this mess. The manager replied "Why, we have never had one blow up" I politely got the hell out of there and refused the offer.
The funny thing is, a buddy works there building the things, and not only did he tell me the Manager was full of it, ( a cursory examination of the design was all I needed ) they still build them the same way today.

 

[Sparweb]

Failure of management to percieve poor technical skills in their personnel means that the responsibility passes up yet another notch. That's the chain of command.

As one goes up in authority, so too, does one take on greater responsibility. If the technical work is shoddy, the tech's supervisor must be able to recognize that, because he's responsible for the tech's work. If the manager is sloppy or oblivious, then HIS supervisor (department head or whatever) must take action because the manager will allow gross errors to slip by that could endanger the company - or lives.

By blaming only the tech that made the mistake is like picking a "fall guy". The poor sod might not have had the experience or training to percieve a flaw, or was doing what he was told. Someone a bit removed from the minutiae can be in a better position to find design flaws, therefore should be capable of doing so. I have personally found myself engrossed in the process of selecting just the right rivet and just the right bend angle when the boss strolled up and pointed out a fundamental problem with the design.


STF

 

[JOM]

funnelguy,
(BTW, what's the story behind that nickname? Do you collect funnels? Do you look like one? )

The account of your mishap is chilling. Here's another similar story with a mos different outcome. Where's the management failure? It's there, it just has to be brought to the surface.

STORY:

The company manufactures paper. A contractor was carrying out repairs on a reel splitter on a paper machine. The reel splitter cuts discarded paper rolls and feeds the waste to a repulper. The reels are lifted by a bucket (like a front-end loader bucket) to a guillotine – both are hydraulically powered. The bucket was in a raised position to permit
testing of the guillotine.

The contractor discovered an electrical fault and called an electrician for assistance. The electrician apparently completed his task and was seen to move towards the end of a line of paper reels.

The contractor went to a control panel and lowered the bucket. Unbeknownst to him, the electrician had returned to the reel splitter and was crushed when he became entrapped between the bucket and a 1.5 metre diameter (3 tonne weight) reel of paper. He sustained fatal injuries.


Cheers,
John.