April 17, 2005
Paris Airport Collapse - Final Report
I noticed on Friday that the unsustainable Amtrak railroad's flagship Acela service had to be shut down because the Canadian made brakes on these French made trains were fracturing. (Affecting 300 of the 1200 total brake rotors in the Acela fleet.) I see today that Amtrak COO Bill Cosbie, errr, Crosbie, has several estimates for when the trains might resume service ranging from next Wednesday to two months or more.
My immediate reaction on Friday was, "another fine example of French engineering" which reminded me I hadn't heard any more news about the Paris airport terminal collapse since my blogs on the subject last August and May. So I Googled news for "paris airport terminal collapse." Only 4 hits - clearly not a hot news story any more - and one of them is actually relevant.
Paris Airport Roof Structure to be Replaced, from the 'Insurance Journal' website. The story cites a February 2005 report by an expert commission that concluded, "both structural and design faults were the underlying cause of the collapse." I suppose that means structural construction and structural design.
The story also references a February 16 IJ article on the release of the report, which I found in the site's International News Archive. From that story we find this summary:
An investigative commission under the direction of Jean Berthier, engineering Professor at France's Ecole Nationale des Ponts et Chausées, concluded that the building's structure had been fragile from the outset. It then progressively degraded under use - principally from the side walkways - to the point where the structure gave way.Berthier's report pointed to four connected causes: 1) insufficient or badly positioned structural steel; 2) lack of mechanical "redundancy," in that the stresses were concentrated and could not be shifted to other structural components; 3) concrete beams that offered too little resistance to stress and use, and 4) the positioning of metal supports within the structural concrete.
OK, so there's no direct mention of square windows in a concrete arch leading to its failure, but this sure seems to be an indictment of the structural design. A design which relied upon steel, and apparently not enough of it, to help concrete do a job that it wasn't suited to. If I ever get my hands on the complete report I might find more details to support my original hypothesis.
Another connection I tried to establish was between this engineering failure and the Socialist system in France that advances individuals into positions of authority for reasons other than merit. That charge is echoed by a commentor to the IJ piece from February.
After giving his own explanation for the failure (Structural Instability of the Euler Buckling or P-Delta Effect) Dr. Michael P. Hogan added, "Someone should check the other buildings this designer has "created" - before the next insurance claim." This was answered by a Dr. Doug Shamanon with, "It is clear that you just have a personal axe to grind, this is unprofessional on your part." This really got Dr. Hogan fuming:
"I do have an axe to grind Dr. Shamanon.There is a pattern of incompetence in the design of major public structures.
The Inquiry into the collapse of de Gaulle Airport says the structure was too weak to sustain itself and progressively collapsed.
We had a similar problem here in Canada when the roof of the Olympic Stadium in Montreal was not able to support itself.
We all pay insurance and we all pay for and use public buildings.
Is it too much to expect them to be able to stand up under their own weight?
When Architects stop playing God I will no longer have an axe to grind.
Too many times, the Structural Engineer (i.e. me) has been told by the Architect - "Make it work or I will find another Engineer". Architect's audacity compromises the safety of structures and there seems to be no check on them.
In the case of de Gaulle Airport Collapse, M. Andreu is BOTH the Architect and Engineer of record."
The only thing I would add is if you're going to make one man both the architect AND the engineer on your project you'd better make damn sure he knows what he's doing. If your choice happens to be a government official then by all means, don't just go by his "credentials."
I think we agree JohnGalt on Architects, or as my brother (a civil engineer by profession) would call them "Art-chitects" where the emphasis is on the art of the structure rather than the engineering. The architect does often call the shots as Dr. Hogan laments. Look at any major civil project and you will see the architect's name listed, often as designer. Rarely do you see the same for the structural engineer. These high profile guys do call the shots, and often stretch the boundaries of engineering to make their mark with their designs. All fine if as Dr. Hogan states the engineering is valid, for on civil projects all taxpayers are financier's of the project. Civil engineering is also the only branch of engineering that does not routinely perform testing on the actual product. Everything else from your dishwasher to your car to the aircraft you fly on undergo extensive testing to prove the design and verify the structural analysis. Buildings and bridges do not and so should have even more emphasis on rigorous design analysis.
Posted by: Silence Dogood at April 18, 2005 11:53 AMI'm going to try and goad Johngalt into saying something bad about Howard Roark...
Yes, gang, architects can fail but are not the best designs those where form follows function? In the software world, I love to throw out a design because "it's too hard." If you have to make a lot of special exceptions and epicycles and question things every step of the way, your design is bad and needs to be scrapped. Hard for me to imagine that architecture is that different.
Lastly, I will make a Virginia Postrel case. I can imagine structural engineers not seeing the value in design and aesthetics and needing a little push sometimes by a good architect.
"Did you say knives?" "Rotating knives, yes."
Posted by: jk at April 18, 2005 12:45 PMRoark would agree with "the best designs [are] those where form follows function" and so would I.
As for your criticism of (overly conservative) structural engineers, a good architect will encourage him to consider innovative structures but will NEVER threaten him to "make it work or I'll get another engineer." The problem with that attitude is that every engineer competent enough to know it's a bad idea will reject it, until the architect finds an engineer not competent enough. Then we have a disaster. And yes, Roark would agree with this as well.
Try again! :)
Posted by: johngalt at April 19, 2005 2:27 PMNah, I'm beat. Did you see the episode of "Barney Miller" where the suspect calls himself "Roark" and Dietrich caught the allusion and foiled the bomb plot?
Posted by: jk at April 20, 2005 11:56 AM | What do you think? [4]