soft systems modeling

[Wayne Wakeland]

I have asked students to use game theory-based concepts (iterated
prisoners dilemma) to model "tragedy of the commons" situations using
STELLA as the modeling language...with very interesting results. The
more rigorous mathematical foundation (compared to ad hoc TOC models
often seen in the literature) is good in some ways, but tends to impose a
structure that seems artificial in an S-D model.

Wayne Wakeland
hwww@odin.cc.pdx.edu
Adjunct Assoc. Prof.
System Science Ph.D. Prog.
Portland State U.
Portland, OR 97207

[Lutz.Schlange@SGZZ.unisg.ch]

Skipping through the sd archives on the sd website it was interesting to follow
up on the discussions about sd modeling using soft variables that was
triggered by Stever Robbins in December 1994. Following are some short comments.
As a matter of fact, there is a school of thought dealing explicitly with soft
systems modeling. Based on earlier findings in system dynamics, it was mainly
developed by futurists such as Frederic Vester from Munich, Germany. Other
interesting approaches have been developed by management thinkers like Peter
Checkland or Colin Eden from the UK.
Vesters methodology (he calls it sensitivity model) is designed to build
holistic feedback models. They are based on the mental models of participants
in the process thus taking into account qualitative factors in combination with
hard quantitative facts.
I have done a paper on some recent work in progress where I used systems
thinking by applying Vesters methodology (L.E. Schlange, Linking futures
research methodologies. An application of systems thinking and metagame
analysis to nuclear energy policy issues. FUTURES, October 1995). In client
processes we have modeled variables such as public acceptance of nuclear
energy, managements cooperative attitude towards anti-nuclear groups etc.
applying a fuzzy logic technique. Also, we use a fuzzy logic based simulation
tool which is highly transparent for GDS purposes.
By the way, metagame analysis or its latest offspring called drama theory is
explicitly dealing with problems of cooperation and conflict under the paradigm
of irrational choice (the drama theory mailing list can be accessed via
internet: "nisp.ncl.ac.uk:70/11"). To my mind, the quest for modeling human
behavior in interactive processes demands a combination of sd and game
theoretic approaches.

Lutz E Schlange
SGZZ St. Gall Center for Futures Research
_________________________________________________________________________

Lutz E Schlange
SGZZ St. Gall Center for Futures Research
Dufourstrasse 30
CH-9000 St. Gallen
Switzerland
Tel. +41 71 24 28 16
Fax. +41 71 24 13 10
Email Lutz.Schlange@SGZZ.unisg.ch
Internet http://www.unisg.ch/~sgzz/index.html
_________________________________________________________________________

[D.C.Lane@lse.ac.uk]

The comments from Lutz Schlange concerning soft systems modelling and
the use of fuzzy logic in his work is an interseting contribution. He also
mentions the contributions of Eden and Checkland.
Participants might be interested to know that a special edition of the
System Dynamics Review - Volume 10, Number 2/3 - was published in 1994
containing articles by these and other soft thinkers.
My paper introductory paper in this edition attempts a brief survey of
some of the soft approaches and a discussion of how they relate to system
dynamics.

Dr. David C. Lane
Operational Research
London School of Economics

[PKucera@aol.com]

Lutz:

I noted with particular interest your comment about the relationship between
system dynamics and game theory. As a graduate business student a few years
back, I saw some synergy there, but even now Im unable to articulate it very
well.

Id be very interested in knowing more about how others view the connection.
Any thoughts?

Thanks.

Paul Kucera
High Performance Systems
PKucera@aol.com

[Moderators note. Dong-Hwan Kim presented a paper at Tokyo on this subject you can contact him at dhkim@techno.etri.re.kr for more information.]