Re: Seen in New Scientist

I'm not sure specifically what part of this statement you see as
problematic. If you assume the worst, the statement sounds ignorant;
but if they did the analyses correctly, it might be reasonable.

It seems feasible to estimate with, say, 95% confidence, the maximum
wind speed likely to occur within a given 100 years. Knowing that,
the statement seems to make sense, more or less -- at least it defends
the safety of the buildings over their expected design life of 60
years.

One quible might concern imprecision of the term "unlikely" above; but
hopefully this refers to some usual convention, like a less than 5% or
1% chance.

That raises the issue of how much confidence one wants or needs here.
Let's say "unlikely" here means 5%. That means there is a 5% chance
of there being wind speed within 100 years that would threaten the
building. (Would you want to work or live in the building?)

What gets very vague is whether the excess 40 years in the predictive
timeframe implies greater safety in the first 60 years. That is, if
there is only a 5% chance of damaging winds in the next 100 years,
presumably there is lower chance in the next 60 years.

But in that case the safest procedure would seem to quantify the
latter exactly.

In addition, and perhaps more importantly, there is inherent
uncertainty in the assumptions of the predictive model to consider.
Presumably these predictions are based on historical data. But to
what extent can we use historical data to predict the next 100 years'
weather (especially given consdierations like global warming)?
Realistically this uncertainty would need to be factored into the
predictions. Considering this uncertainty, builders would need to
apply more rigorous design methods to achieve the same nominal degree
of (e.g., 95% percent of withstanding any possible wind in the next x
years) of safety.

John Uebersax PhD
http://satyagraha.wordpress.org


On Oct 30, 5:26 pm, John Kane <jrkrid...@xxxxxxxxx> wrote:
From the letters section of New Scientist:
“ Over the years I’ve had numerous arguments with engineers and
architects about the basic design parameters they use and have been
told quite seriously that, for instance, “This highrise block of flats
is designed to withstand a wind speed unlikely to be exceeded in 100
years, so since the design life is 60 years we have 40 years to
spare.”

John Kane Kingston ON Canada

.

Relevant Pages

  • Re: Seen in New Scientist
    ... That means there is a 5% chance ... uncertainty in the assumptions of the predictive model to consider. ... Presumably these predictions are based on historical data. ... apply more rigorous design methods to achieve the same nominal degree ...
    (sci.stat.consult)
  • Re: Seen in New Scientist
    ... That means there is a 5% chance ... uncertainty in the assumptions of the predictive model to consider. ... Presumably these predictions are based on historical data. ... apply more rigorous design methods to achieve the same nominal degree ...
    (sci.stat.consult)
  • Re: TomS didnt read the article very closely and Dr Rana argues from ignorance of ID theory
    ... >>"Creation Scientist Challenges Intelligent Design ... > of the origin of life which they call a "Creation Model." ... have noted that Dr. Rana is the "director for science apologetics" (how many ... Some theories, however, sometimes make predictions ...
    (talk.origins)
  • Re: Question for T. Pagano on Modes of Causation
    ... three "modes of causation," law, chance, and design. ...
    (talk.origins)
  • Re: Question for T. Pagano on Modes of Causation
    ... three "modes of causation," law, chance, and design. ...
    (talk.origins)
Loading