Since the times of A.M. Freudenthal, the problem of structural safety begins with the answer to the question, "What is an acceptable risk of failure?" The answer to this fundamental question is as elusive today as it was 50 years back. There have been several proposals and counter proposals (also proposals to counter the counter proposals) for estimating failure probabilities. The basic question of appropriating an acceptable risk of failure has not received the attention that it merits.
A commonly used specification for pegging the design basis earthquake is 10% exceedence probability in 50 years (corresponds to a return period of roughly 500 years). The basic issue of acceptable risk is far from being settled. The fact that several non-engineering issues such as social, political, and economic aspects have a bearing on the definition of an accepted level of risk makes the task of engineers involved in framing design codes much more difficult.
One of the proposals to resolve this issue takes the position that the probability of getting killed in any structure/facility during an earthquake should be equal to, or less than getting killed in an auto accident, or while working. Alternatively, the acceptable earthquake risk is related to risks that all of us are exposed to every day and have accepted in the past [Wiggins, 1972].
Yet another proposal, which has formed the basis of many a building design codes is that the solution to be adopted should be that for which the total cost of construction (including the cost of repairs in the event of damage at present rates) is a minimum [Torroja, 1948].
There are some obvious shortcomings in both the approaches - some engineering, and some philosophical/ethical difficulties as discussed by Grandori.
What is your take on this issue? Let us initiate a debate on this fundamental question of earthquake safety.
[Gandori, 1991] Grandori, G., Paradigms and falsification in Earthquake Engineering, Meccanica, Vol. 26, 17-21, 1991.
[Torroja, 1948] Torroja, E., Load factors, Journal of the American Concrete Institute, Nov. 1948, 567-572.
[Wiggins, 1972] Wiggins, J.H., The balanced risk concept, new approach to earthquake building codes, Civil Engineering, ASCE, August 1972, 55-59.