Hi Bin,
I'm unsure of how many papers tackle the issue thoroughly, I
think it is rare. And, it seems to depend on the type of building
(if buildings is what you're asking about).
For publicly available buildings, like libraries and department
stores (which in many ways functions as urban areas in as far as
movement patterns are concerned) I have found high correlations
using space syntax theory in "section", that is, drawing axial
lines in section at a reasonable "eyesight" height (it leads to a
half-stair being one syntactical step, and a full stair being
two, et cetera). This translates into depthmap as well, you have
to "fake" syntactic steps that correspond to the section/3d
syntactic steps all after sps theory. It gives correlations up
towards 75-90%. Seems to be similar in urban environments (like
the multi-planar "slussen" in Stockholm), but not quite as true
for other buildings where social organizations are more defined
(and are more decisive for movements) such as offices, if I
remember correctly from what I've heard and read of my colleagues'
works.
However! This method does not work for angular analysis, for
instance, in depthmap, since the idea of creating syntactic steps
by closing a connection and then connecting past it with a
superlink (if it's still called that?) makes no difference in
angular analysis but does in a standard integration analysis.
I didn't publish these findings in a separate paper, I think.
Perhaps I should, but I think a description of the modeling
method can be found in either of my thesises
(http://www.arch.kth.se/sad/publications_index.htm), with more on
depthmap in
the newer (actually, I changed method in the newer to closer
correspond to an axial-line modeling of the stairs, which raised
the correlation of the vga-analysis significantly).
The short answer is "think in 3d how many spaces, minimum, you
have to pass to pass the stairs and find a way to model that in
the model".
For elevators, also, it seems to work better when modeled on a
per-foor basis with a syntactical step representing the doors as
well, and _not_ by modelling the elevator as one space for all
floors. The former tend to make the elevator a longer trip for
one floor, somewhat similar for two floors (depending on the
stair layout and position!) but an increasingly shorter trip for
several, which is also how they seem to perform.
I could maybe fish out the parts I have in my works about the
modeling for you since it's rather short parts in a very long
thesises. If there isn't better work done it out there somewhere
that I've missed as I haven't been able to keep that up to date
for a while.
/dk
Bin Jiang wrote:
Hi, I once came across papers on modeling movement between
floors. Can
anyone there point me to related papers or reports?
Thanks and cheers.
Bin
--------------------------------------------------------
Bin Jiang
Division of Geomatics
Dept. of Technology and Built Environment
University of Gävle, SE-801 76 Gävle, Sweden
Phone: +46-26-64 8901 Fax: +46-26-64 8828
Email: [log in to unmask] Web: http://www.hig.se/~bjg/
--------------------------------------------------------
NordGISci: http://www.hig.se/~bjg/NordGISci/
ICA Commission: http://www.hig.se/~bjg/ica/
******************************************
Daniel Koch
Architect, Ph.D. Student
KTH School of Architecture, Sweden
+46 8-790 79 79; www.arch.kth.se/sad
-------------------------------------------------
This mail sent through IMP: http://horde.org/imp/
|