System Dynamics in Water/Wastewater Mgmt

[jm62004@Jetson.UH.EDU]

At 09:04 AM 3/8/97 +0100, Roberto Vacca Wrote:
>Dear Jaideep: I think there are serios fallacies hidden n the paucity of
>data acceptance. At ISIS, Rome, Italy we use SD, but never alone. After
>major drives to get data & assess their quality, we use lgistic
>substitution models [based on Volterra 3-parameter logistc equtions] and,
>where appropriate i.e. in regional o urban models, also econometric
>Input-Output models. We produced models of Rome city &are workig now on
>land use-transportation-energy-emissions-dispersion-environment on
>contract to European Commission in Brussels.
>Concerning data & xplanations, Galileo wrote:"whatever our senses and
>experience indicate, has to be put above any discourse - even if the
>latter
>sounds well founded. Look up in literture papers by C.Marchetti &
>N.Naikichenovich.
>Best. Roberto Vacca
>MC4634@mclink.it

Dear Roberto

Thanks for your reply. Your points are well-taken - ideally we would like
to have the best statistically valid parameters in SD models, but that is
not always possible. I view SD as a (VERY) extended way of back-of-the
envelope calculations - which are necessary because we have to deal with a
big and complex reality all the time, and so we cant always wait for the
best estimates. I really wonder if in social systems there can ever be any
fixed parameters - much of neoclassical economics and quantitative modeling
acts as if you can have E=mc^2 type equations to cleanly describe human
behavior. I doubt it, so we take a middle of the road (muddling-through
approach) between the verbal-emotional-intuitive and the
mathematical-analytical-objective.

Thanks for your valuable comments,

Regards

Jaideep
jm62004@Jetson.UH.EDU
****************************************

Jaideep Mukherjee, Ph. D.
Research Associate
Department of Industrial Engineering
University of Houston
4800 Calhoun Road
Houston, TX 77204-4812

Phone: 713 743 4181; Fax: 713 743 4190
****************************************

[antonio barron]

REPLY System Dynamics in Water/Wastewater Mgmt (SD0761)
Jaideep wrote:
> (To Mr. Roberto Vacca):
Your points are well-taken - ideally we would like
to have the best statistically valid parameters in SD models, but that
is not always possible.

Dear Jaideep:

I agree with you in this introduction of your message.
When you try to build models to explain how become the behaviour of a
collective (for instance when you are making essays to describe the
introduction of new products into the MARKET) you are thinking that is
dependent of a lot of factors: the own (innovation) behaviour of
persons,the price of goods, the quality, the publicity, the marketing
strategies, the cover or distribution channels,... too much, and
moreover you lack of historical information, if the product or service
is new !

So, what can we do ?
We can throw in the towel, and go to other thing. What a pity !
Or we can collect the experience of other authors, that tried to explain
similars systems, to read the expertises opinions in this MARKETS,
to make opinion poll, market research, trying to get the best approach
to the phenomena, and it is possible as mathematical formulas.
Perhaps you will get a better result than if you only see the glass
ball, and with the new information (and with the help of powerful
analytic tools) you can increase the accuracy of such a model.
Finally: I will insist another time about the importance to listen to
the expertise (the person that better knows the system to be modelled).
That will contribute to start in a good way !

Example about "A simulation model for partially substituting
telecommunication services" .You can consult in the webs:
http://web.mit.edu/jsterman/www/SD96/home.html or
http://ourworld.compuserve.com/homepage ... s/sd96.htm

Regards.
Antonio Barron

***************************************************
Antonio Barron
Superior Engineer of Telecommunications (EIST of Madrid)
antonio.barron@telefonica.es
Telefonica de Espana S.A.
C/Valverde 2,piso 12
Madrid 28004
Phone 34-1-5844791 office
Fax 34-1-5231706 or 5232350 office

Private phone: 929 131931
E-mail: 100407.2170@compuserve.com
***************************************************