SD Models from Written Text???

This forum contains all archives from the SD Mailing list (go to http://www.systemdynamics.org/forum/ for more information). This is here as a read-only resource, please post any SD related questions to the SD Discussion forum.
Locked
"Jay W. Forrester"
Senior Member
Posts: 63
Joined: Fri Mar 29, 2002 3:39 am

SD Models from Written Text???

Post by "Jay W. Forrester" »

There has been an interesting discussion exchange on formal methods
to derive models from written text. The effort is probably worth
pursuing for the insights it will yield about process.

However, in any rather complicated modeling project, only a very
minor fraction of the relevant information is to be found in writing.
Overwhelmingly, the richest and most extensive information lies in
the mental data bases of people working in the real-life situation.
For the sake of illustration, we might say that there is a million
times as much information in peoples heads than in writing, and in
turn the written data base is probably a million times more useful
than he numerical data base. Should anyone doubt, consider that
everyone in a company leaves and is replaced by a person without
experience in that situation. The replacement is told to read the
material and look at the numbers in the office and carry on--total
chaos would result because the organization is run on the basis of
experience, apprenticeship, and "being there."

Perhaps the emphasis should be placed on interview processes, what to
look for in listening, what kinds of questions to ask, and how to dig
below the surface for information that is not initially forthcoming.

Creating a model is creating a theory of behavior of a situation in
the real world. There has been no clear guidance that guarantees the
creation of theories. There are no ten rules that will promise
receiving a Nobel Prize in physics. Modeling need not be so great a
challenge, but it is of the same class of activity.

I would like to see a discussion on how to condition and sensitize
people to the process of creating good models.
--
---------------------------------------------------------
Jay W. Forrester
Professor of Management
Sloan School
Massachusetts Institute of Technology
Room E60-389
Cambridge, MA 02139
From: "Jay W. Forrester" <
jforestr@MIT.EDU>
Jay Forrest
Junior Member
Posts: 12
Joined: Fri Mar 29, 2002 3:39 am

SD Models from Written Text???

Post by Jay Forrest »

Two observations seem pertinent.

1) There is a difference between deriving a model from text and developing
a model in an interview process. This thread started (as I recall) with
George Richardson suggesting that it would be interesting to have expert
modelers take the same information (text) and observe/analyze the
process
esults.

2) Effective interview methods and information surfacing methods are a very
different topic.

I personally know of no studies/books regarding systems model development
from fixed information bases. This would seem to be new territory for SD
with significant potential value.

On the other hand, the literature of mental model formation, elucidation,
facilitation, etc. is substantial. While not specific to SD, two recent
books from England offer valuable insights for systems thinking/science
practitioners. They are "Systems Approaches to Management by M. C. Jackson
and "Systemic Intervention" by Gerald Midgley. Jackson provides a wonderful
overview of systems approaches, exploring the strengths and weaknesses of
each. Jackson also suggests three categories of model building environment
- individual, group (open and communicating), group - coercive (dominant
individual, anatagonistic environment, dominating boss, etc). Jackson
actually has six categories, but these three are enough to suggest that the
answers to Jay Forresters remarks vary with the nature of the group
involved. Midgleys book suggests a philosophy and methodology for
integrating various systems approaches in real world interventions.

My personal experience is consistent with Midgleys. I would suggest that
no singular systems approach - or method for developing a systems model -
will always be successful - in all environments. Skilled model developers
need a range of methods and alternatives for surfacing the appropriate
insights and information; for communicating and confirming the information
acquired; and for assimilating, validating, and confirming (with the
client) the model developed; for identifying the implications of the model
developed; and for communicating the implications of the model developed to
others. IMO the group, their dynamics, their abilities, interests and time
available, and the skills and knowledge of the facilitator combine to
minimize the practicality of predefined, branching expert approaches. (I
would suggest these work best in environments where the
environment,alternatives, and decision points are well defined. I dont
think we are there with facilitation and human systems!)

Of the two books, Jacksons is more rigid and pragmatic and would probably
appeal more to consultants. Midgley is far more philosophical. I would
think Midgleys book would be more appealing to skilled and trained
facilitators.

Hope this helps!
Jay Forrest
From: Jay Forrest <jay@jayforrest.com>
Locked