Sunday, November 07, 2004

waiting for disaster

The recent attack on the World Trade Center and the Pentagon leads to the question of why there were no effective measures in place to protect against it. The type of attack was not novel. In fact there have been popular fictionalized accounts of the use of airliners as weapons. There were many warnings that a terrorist attack could be expected very soon. There had also been many well-publicized tests of the airlines' security systems which showed that they were easily overcome and weapons could be carried onto aircraft by passengers. There was also ample warning that terrorists would not hesitate to engage in suicide attacks. These attacks are common in the Middle East and have even been used against US military targets there. Now, in the clear light of hindsight, it is obvious that the improvements in airline security are necessary.

But why was it necessary to wait for disaster?
Mankind seems to be nearly incapable of taking steps to avoid easily anticipated disasters. If we consider a few types of disasters and recall the history of attempts to avoid them the problem is clear.
Having fire escapes in buildings, adequate lifeboats on ships, dead man controls on trains, and many other disaster avoidance requirements only came into use in response to major disasters. This was true despite the fact that these disasters are very easy to anticipate. Why was it necessary to experience a tragic loss of life before taking precautions that seem like simple common sense?

Perhaps the cost of the precaution seems more important than the potential disaster, which might not happen.

Even after new laws or building codes have been adopted, there is still widespread evasion often with official complicity. For example, in Florida, hurricane Andrew revealed that much of the housing that was destroyed did not have reinforcements required by the building codes that were in force when the houses were constructed. The fact that the houses were inspected and passed indicates official complicity or at least incompetence.

There is also a common tendency to design structures in a way that minimizes cost but does not prevent failure due to circumstances that are considered unlikely. The principle that the designers do not observe is that if the consequence of failure cannot be tolerated then, even if the probability is low, the possibility of such failure must be designed out.

The Titanic sank because its designer thought that only a small number of compartments could ever flood at the same time and therefore the internal bulkheads, which separated the compartments, did not have to extend all the way to the deck.

The World Trade Center towers collapsed because the supports that attached the floors to the vertical beams were strong enough to support the weight of the floor and it's contents but when the supports of one floor failed due to fire, the floor below was not able to withstand the force of the floor above dropping onto it. The floor beams could have been strong enough to support any number of fallen floors since the vertical columns that they attached to were strong enough to support all the weight of the building above them.
The designers of the World Trade Center either did not anticipate this failure mode or thought that it could not happen. Perhaps they thought that the sprinkler system would prevent the failure of the floor beams by controlling any possible fire. There were narrow stairways, not enough of them, and they were not sufficiently protected from smoke.

Of course now, after the disaster, we may hope that any future tall buildings will be designed to eliminate this particular failure mode.
There are probably now many similar disasters waiting to happen. Let's hope that we have learned enough to avoid at least some of them.