REPLY Separate Professional Conference (SD6290)

[System Dynamics Mailing List]

Posted by  martin at utalca.cl

Hi,

I'm at the same time fascinated by the last weeks' discussion and worried about it.

My short (since 2003) personal background as an academic does not allow me to
talk much about the professional aspects of system dynamics.  However, I
believe there are at least two populations inside the SD community, and each has
its reason to be.

In the professional area, there are people who use SD in their work, others
who offer consulting services and the developers of SD software.  Maybe those
who use it in their work have too few time for developing full-blown simulation
models each time; however, some basic insights about stock-flow relationships
and feedback (with polarity) are still helpful, as long as one is not
over-confident in intuitions.

I understand that SD was developed in the first place to just do this: overcome
complex dynamics problems.  I also believe it is a noble motive (regardless of
if you do it for financial profit or other reasons).

But I believe there is still a place for academics (I hope so).  For instance,
some of them are working on the loop dominance, which has to do with how
systems behave.  Other academics are interested in how humans think about
complex dynamic systems, how exactly we fail and how to help.  Still others
prefer scientific theory building over solving concrete problems.  I do hope
that these and other academic interests have their space in SD.

Just in this sense, "Divide and rule, a sound motto. Unite and lead, a better
one." (said Goethe).

There have been different comments about the conference.  Practitioners may
wish more time to spend around their subject area (maybe according to SIGs). 
But also academics have wondered if there could not be more about their kind
of subjects; once a friend of mine (Camilo Olaya from Colombia) thought the
threads outline in Business Dynamics (John Sterman, 2000) would give a nice
architecture:

Theory
-	theory of nonlinear dynamics and complex systems;
-	agent-based modeling;
-	mental models, dynamics decision making and learning;
-	organizational and social evolution.
Technology
-	automated mapping of parameter space;
-	automated sensitivity analysis;
-	automated extreme condition testing;
-	automatic, interactive parameter estimation, calibration and policy optimization;
-	automated identification of dominant loops and feedback structure;
-	automated help;
-	visualization of model behavior;
-	linking behavior to generative structure;
-	data implementation;
Implementation
-	communicating modeling insights;
-	improving group modeling;
-	speeding the process;
-	integrating modeling methodologies;
-	creating managerial practice fields;
-	assessing outcomes
Education


Wouldn’t this way to organize help to have a conference for both, practitioners
and academics?

I also wondered if there could be a clearer distinction between different types
of papers to be presented:
-	conceptual papers that are logically coherent but not yet empirically tested;
-	in-progress papers about projects that are in the pipeline and present
interesting practical problems (about experimentation, for example);
-	finished projects that report final results and insights.

The research sessions were a step in this direction; however, the
time-per-speaker was much too short to have an interesting discussion (better to
be in a poster) and also the allocation did not always do justice to the papers’
nature.

I think that for what concerns diffusing SD to other groups of people may be
done by going to their meetings and sending manuscripts to their journals.  And
we shall be patient we cannot march faster than the music.

"Saludos",

Martin Schaffernicht
University of Talca
Talca - Chile
Posted by  martin at utalca.cl
posting date  Wed, 21 Feb 2007 12:23:01 -0300