Dynamical Systems and System Dynamics ??

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
Luis Gutierrez
Junior Member
Posts: 6
Joined: Fri Mar 29, 2002 3:39 am

Dynamical Systems and System Dynamics ??

Post by Luis Gutierrez »

Hello Esteban,

""System Dynamics"" is a method to study the structure and behavior of
dynamical systems. There are many other methods, but ""System Dynamics""
uses a (by now) fairly standard set of symbols for developing system
diagrams and simulation models. The system diagrams (usually called
""causal-loop diagrams"") depict a hypothesis as to what are the feedback
loops that generate the observed behavior over time. Then, using rate
and level equations (or, in mathematical terms, zero and first order
difference equations), the simulation model is built, and there are a
number of simulation languages to facilitate the process of formulating
the equations and running the simulations.

Generally speaking, if you see graphs of system behavior over time
(based on actual and/or simulated data) *and* you see diagrams with
directed arrows that link system variables as feedback loops -- then you
are looking at a ""System Dynamics"" analysis of a system or process. If
you don't see behavior-over-time plots and feedback loop diagrams --
then you are looking at some other kind of analysis.

To see some examples pertaining to various kinds of systems, visit the
Ossimitz System Dynamics Mega Links Directory:

http://www.uni-klu.ac.at/~gossimit/linklist.php

Do you have any formal training in ""System Dynamics"". There are some
online tutorials, but this is not the kind of analysis that you can
learn easily on your own. I was trained by Willard Fey, who was in the
original group under Jay Forrester at MIT.

Hope this is helpful.
Luis
From: Luis Gutierrez <LTG1979@attglobal.net>
Paul Newton
Junior Member
Posts: 2
Joined: Fri Mar 29, 2002 3:39 am

Dynamical Systems and System Dynamics ??

Post by Paul Newton »

Hi Esteban,

My impression is that, mathematically, dynamical systems and
system dynamics are the same thing in the sense that both are
dealing with change through time. However, the usual system
dynamics practice is to model in continuous time, that is, with time
unfolding continuously, whereas, I have found threads of dynamical
systems practice using both continuous and discrete time
formulations.

Although alike mathematically, there are other aspects of SD and
DS that seem to me to be somewhat different.

1) SD typically focuses on changes through time resulting
from the interactions of feedback loops (feedback causality),
whereas, although feedback is usually present in DS models, I
don't think DS emphasizes feedback to the degree that SD does

2) SD practice focuses on identifying operational causal
relationships between variables, whereas I think DS practice is
mostly silent (doesn't care) on whether relationships between
variables should be causal (structural) or correlational (behavioral).

3) Another difference may be the emphasis in SD on the
modeling process (e.g. in SD literature by John Sterman, George
Richardson, Andrew Ford, Khalid Saeed, Barry Richmond, etc...). I
haven't seen a modeling process emphasis in the DS books I have
seen.

4) Another difference may be that SD emphasizes policy
design (policy engineering) as a primary goal of modeling. I don't
think this emphasis is present to the same degree in DS practice.

And there are probably many other differences. I'll be really
interested to see how others respond to your question.

Paul Newton
Instructor
Cornell System Dynamics Network (CSDnet)
Department of Applied Economics and Management
407 Warren Hall, Cornell University, Ithaca, NY 14853
607-255-5230
pcn4@cornell.edu, or
PaulNewton@StewardshipModelng.com
http://www.CSDnet.aem.cornell.edu
Weaver, Elise A
Junior Member
Posts: 5
Joined: Fri Mar 29, 2002 3:39 am

Dynamical Systems and System Dynamics ??

Post by Weaver, Elise A »

Hello Esteban,

Your posting interested me, because I am always interested in overlaps
between SD and psychology. I looked on the web, and it appears that this
field you mention represents a relatively new movement among social
psychologists looking at the time course of behavior.

These are suggested texts according to the web:
Vallacher, R. R. & Nowak, A. 1994. Dynamical Systems in Social Psychology.
San Diego : Academic Press.
Guastello, S. 1995. Chaos, Catastrophe, and Human Affairs: Applications of
Nonlinear Dynamics to Work, Organizations, and Social Evolution. Hillsdale:
Lawrence Erlbaum Associates.

It appears that system dynamics and this area of dynamical systems in social
psychology share common mathematical roots. SD has been applied broadly, but
not so much to psychology (which traditionally records behaviour in time
snapshots rather than time series). So that may explain the separate
literature streams.

A couple of years ago, I tried to gather some mathematical psychologists to
look at the overlap, but was limited by costs to only inviting a few speakers
from psychology to speak to the system dynamicists. I think there is
tremendous potential for overlap and discussion if we could only figure out
the logistics.

Another overlap is in cognitive engineering such as described in books like:
Jagacinski, RJ and Flach, JM (2003). Control Theory for Humans: Quantitative
Approaches to Modeling Performance. Mahwah, NJ: Lawrence Erlbaum
Associations.

Elise A. Weaver, Ph.D.
From: ""Weaver, Elise A"" <
eweaver@WPI.EDU>
Assistant Professor of Psychology/System Dynamics
Department of Social Science and Policy Studies
Worcester Polytechnic Institute
100 Institute Road
Worcester, MA 01609
Yaman Barlas
Member
Posts: 44
Joined: Fri Mar 29, 2002 3:39 am

Dynamical Systems and System Dynamics ??

Post by Yaman Barlas »

Hello Esteban;
Here is one way of characterizing the relations between dynamical
systems and system dynamics: system dynamics is the name adopted by a
community who does work on dynamical systems of certain type (policy
problems of socio-economic or strategic management nature) with a
special focus on underastanding the feedback causes of the dynamics of
concern (or to choose problems that have significant feedback
structures). This particular focus also (naturally) created a set of
specialized tools (causal loop diagrams; stock-flow modeling,
integration emphasis, special validtity tests, interactive gaming, group
modeling, etc).
So, 'dynamical' systems is more general than system dynamics -as WE use
the term. Dynamical systems just means dynamic systems represented by a
set of differential equations. (The origins is somewhat more specific,,
meaning 'dynamic systems in engineering or natural sciences).
Many dynamical system problems are system dynamics problems, but then
again many are not. All sd problems are at least implicitly ds problems
by nature.
I think we all tend to exaggerate the disciplinary boundaries that stem
from terminology and tools. It is more productive to emphasize
terminology about he content/substance of the work (rather than about
methodology/tools).
I try to 'mix' methodology terninology as much as I can. I sometimes use
system dynamics; sometimes dynamical systems; sometimes dynamic systems,
sometimes dynamic systems approach, sometimes systems science/theory,
etc... in the SAME talk, on purpose.
best wishes
Yaman Barlas
From: yaman barlas <
ybarlas@boun.edu.tr>

---------------------------------------------------------------------------
Yaman Barlas, Ph.D.
Professor, Industrial Engineering Dept.
Bogazici University,
34342 Bebek, Istanbul, TURKEY
Fax. +90-212-265 1800. Tel. +90-212-359 7073
http://www.ie.boun.edu.tr/~barlas
SESDYN Group: http://www.ie.boun.edu.tr/labs/sesdyn/
-----------------------------------------------------------------------------



Esteban Ribero wrote:

>
>
Khalid Saeed
Senior Member
Posts: 79
Joined: Fri Mar 29, 2002 3:39 am

Dynamical Systems and System Dynamics ??

Post by Khalid Saeed »

Colleagues,

In terms of practice, I have following observations to make:

1. Model boundary:
In case of dynamical systems modeling, the model boundary is often based on
a specific complex pattern of deterministic behavior, such as bifurcation,
chaos, oscillation, suspected to exist in a system.

In case of system dynamics, we often spend considerable time constructing
custom reference modes addressing complex patterns subsuming growth,
oscillation and policy resilience, which lead us to the model boundary.
Personally, I'd like to see us slicing complex patterns into their generic
components and basing our model boundary on the multiple patterns
encountered in a given problem, but that seems to be a subset of the
general practice.

2. Solution/understanding
In case of dynamical systems modeling, the explanatory process resides in
the solution expressed in terms of initial conditions, parameters,
mathematical functions and time, not in terms of causal relationships
between variables. No variables appear on the right had side of a formal
solution. Complete formal solutions in such forms can be found for simple
models with linear relationships. Solutions to more complex (higher order
and nonlinear) models are often found through repetitive computer
simulations that can identify the parameter ranges over which certain
behavioral characteristics may appear. The solution process in both cases
attempts to cover behavioral implications for all possible parameter sets,
but does not provide a causal explanation of the behavior.

In case of system dynamics modeling, solution manifests in explaining the
dynamic behavior of the model in terms of the changing dominance of its
underlying feedback loops, although behavioral implications for all
possible combinations of parameter sets may not be known. Computer
simulation experiments are the means to getting to the understanding in
terms of the influence of the dominant feedback loops on the behavior and
how this influence can be changed, but there is no prescribed way to
conduct simulation experiments and represent their results, hence modeler
skill is the key to arriving at meaningful results.

I am lately seeing solution related simulations similar to dynamical
decision systems practice being offered in our modeling practice too. I
think this is useful provided an explanation from a feedback perspective is
also given.

Khalid Saeed
From: Khalid Saeed <saeed@WPI.EDU>
Stu Eldred
Newbie
Posts: 1
Joined: Fri Mar 29, 2002 3:39 am

Dynamical Systems and System Dynamics ??

Post by Stu Eldred »

> who does work on dynamical systems of certain type (policy problems of
> socio-economic or strategic management nature) with a special focus on


Very well said. I know that I propose much better solutions when I focus on
the problem, not on the tool.

You are probably familair with the phrase: ""When the only tool you have is
a hammer, every problem begins to look like a nail.""

Stu Eldred
Acton, MA, US
Joel Rahn
Junior Member
Posts: 15
Joined: Fri Mar 29, 2002 3:39 am

Dynamical Systems and System Dynamics ??

Post by Joel Rahn »

I hesitate to stir up this sort of hornet's nest but field theories
(electro-magnetic, acoustic,...) do not use material stocks and flows
and they can represent such phenomena as the Doppler effect. However
quantized versions (photons, phonons,...) use flows and densities (i.e.
numbers per unit volume, area).

These theories allow for 'action at a distance' but it is limited by the
speed of transmission -no problem there for SD. The truly weird feature
about quantum mechanics is 'instantaneous action at a distance' -let's
not go there in this discussion, please.

I don't know of any examples in which the classical or quantized field
theories have been translated into SD models.

Joel Rahn
jrahn@sympatico.ca
Locked