Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re: Housing Performance Objectives

[Subject Prev][Subject Next][Thread Prev][Thread Next]
In a message dated 97-12-02 14:45:19 EST, you write:

<< Only major changes in the status quo will cause policymakers to rethink
this
 issue now. >>

I definately see the two "camps" described by Fred. (i.e. life-safety and
life-safety+damage control). However, I see this as part of a bigger issue:
how can we objectively debate code provisions while each debate is based upon
differences in philosophy?  Also, even when  the parties share the same
philosophy, there's another big question, how much regulation is enough?  

In  order to have a rational decision making there must first be an  agreed
upon measure of effectiveness.  In  other words, the processess of finding
common ground between two differing code groups will not  be consistent
unless held against some measuring stick of public utility. The textbook
measure of public utility has been a probabilistic determination of the e.
expected value versus expected cost.  These  methods have the unfortunate
side-effect of dollers per statistical human life  somewhere in  the
equation, which is a no-sell politically.   Those who embrace LRFD may point
out that probabilistic methods are already incorporated into the UBC in
Strength Design, however as a whole the  model codes are lightyears from
having a  uniform factor of safety, which would occur in a completely
rational approach to the code.  Try to figure what the  factor of safety is
in the various fire-safefy provisions!  I look forward to the  day  when all
regulations are based on a candid, (life-cycle) assessment of expected values
and that only efficient life and property  saving investments were codified.
(Actually I look forward to just having consistant load factors in strength
design :>)  As it is, a scientific risk-assessment is twice removed from the
decision making process because the process does not have a completely
rational objective.   

My  intent is not to "trash" the code, just to point out that there will
probably always  be inconsistencies.  My view is that the code  will not
 have a uniform factor of safety  until: 1)  we  decide what is an acceptable
level of risk (%) for life and property, 2) probabilistic methods are made
politically acceptible, and 3) the code becomes truly performanced based.  


Just my 2 cents.

Tom VanDorpe