Need a book? Engineering books recommendations...

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

Re: "Analytical model"

[Subject Prev][Subject Next][Thread Prev][Thread Next]
But, it is an analytical model.

It is a model used to ANALYZE an idealized structure.  Yes, once your reach the point of getting to an FEA model, it will basically be a numerical analysis, but it is still an ANALYSIS.

Now, I do agree that there is a danger posed from BIM (and other computer analysis methods) in that a lot of people forget that part of the analysis is going from your REAL structure to some IDEALIZED structure that can be analyzed.

While Richard does have a point, he also glosses over the fact that as structural engineers NONE of us is actually analyzing and designing the real structure directly.  We are all analyzing and designing some idealized version of the structure that, if we are doing our job correctly, does a reasonably good job of representing the real structure.  For example, we approximate a beam with double angle connections at both ends as being equivalent to a simply supported beam with a pin support and a roller support.  We neglect the fact that such connections actual DO have some resistance to rotation, but it is so negligible that we can ignore it.

So, I agree that there is an issue that some engineers (frequently younger) get so wrapped up RUNNING the model that they may not understand how to create the model from the real structure such that the idealized model will reasonably represent the real building.  And BIM can certainly contribute to that problem.

But, I will also point out that the it is an analytical model.

It sounds like your issue is that the BIM software is the one doing the "idealization" not the engineer.

Regards,

Scott

On Oct 1, 2011, at 5:14 AM, Alexander Bausk wrote:

As you may know, the Revit world wildly enforces the term "analytical model" which means essentially the FEA model generated from a BIM "physical" model.

Is it the same pain to see for you as it is for me? FE Model isn't analytical in any way, it is numerical.
Okay, "analytical" may be a homonym which means both, ehm, "calculational" and "such that has a closed-form solution", but doesn't it pose an actual threat to understanding of the subject,
especially by younger generation?

Oh, and I had a very hard time explaining the difference to architectural folks at the recent Autodesk Russia Forum in Moscow. Something clearly needs to be done here.

--
Alexander Bausk
Civil/Structural design & inspection engineer, CAD professional
MSc Structural engineering, Ph.C. Engineering
http://bausk.wordpress.com
ONILAES Lab at PSACEA
Dnipropetrovsk, Ukraine
Tel. +38 068 4079692
Fax. +38 0562 470263
bauskas(--nospam--at)gmail.com