QUERY Structural changes and validity

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
Monte Kietpawpan <kietpawpan@
Junior Member
Posts: 12
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by Monte Kietpawpan <kietpawpan@ »

Posted by Monte Kietpawpan <kietpawpan@yahoo.com>

Dear all,

Structual validation is key to the success of model development.
However, we have no test to determine whether a given model structure
is stable over the time horizon of interest.

Basic mechanisms may at best represent the ""past structures""
capable of generating the ""past behaviors"". No existing validity
tests serve to check whether the relationships in the real system
will not significantly change in the future--within the time horizon.
If the real structure changes, the model of past behavior would
lack the ability to generate the future behavior. It is therefore
hard believe that SD models are theories, unless the structures
of the aspects being modeled are stable.

SD is not effective in dealing with social problems, of which
the dynamics of the system structures are not yet predictable.
Projection of dynamic behavior, to be successful, calls for the
ability to project the dynamics of the structure, the ability that
the oracle of Delphi posses.

MK
Posted by Monte Kietpawpan <kietpawpan@yahoo.com>
posting date Thu, 4 Oct 2007 02:33:59 -0700 (PDT)

_______________________________________________
Monte Kietpawpan <kietpawpan@
Junior Member
Posts: 12
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by Monte Kietpawpan <kietpawpan@ »

Posted by Monte Kietpawpan <kietpawpan@yahoo.com>

Dear all,

Structual validation is key to the success of model development.
However, we have no test to determine whether a given model structure
is stable over the time horizon of interest.

Basic mechanisms may at best represent the ""past structures""
capable of generating the ""past behaviors"". No existing validity
tests serve to check whether the relationships in the real system
will not significantly change in the future--within the time horizon.
If the real structure changes, the model of past behavior would
lack the ability to generate the future behavior. It is therefore
hard believe that SD models are theories, unless the structures
of the aspects being modeled are stable.

SD is not effective in dealing with social problems, of which
the dynamics of the system structures are not yet predictable.
Projection of dynamic behavior, to be successful, calls for the
ability to project the dynamics of the structure, the ability that
the oracle of Delphi posses.

MK
Posted by Monte Kietpawpan <kietpawpan@yahoo.com>
posting date Thu, 4 Oct 2007 02:33:59 -0700 (PDT)

_______________________________________________
George Richardson <gpr@albany
Member
Posts: 20
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by George Richardson <gpr@albany »

Posted by George Richardson <gpr@albany.edu>

The writer of this query suggests that system dynamics can not be effective
in dealing with social problems because we don't know future social system
structure. It's worse than that, I suppose, because we can't even know
current social system structure for sure. As he says, we can not ""validate""
structure, and we have no test to determine if structure is ""stable"" over
time.

If I understand the posting correctly, the implication appears to be that we
should stop thinking about future social problems because we don't know how
current social structure will evolve. Not only should we not model difficult
current societal problems, we should stop thinking about them in any form,
since the social structures we'd assume in our thoughts can't be ""validated.""

I prefer to think that we have to do the best we can to think about societal
dynamics as they move from the past into the future, rather than refuse to
try to apply the best tools we currently have for thinking about complex
dynamic systems.

It helps our courage to remember that ""No model has ever been or ever will
be thoroughly validated. ... 'Useful,' 'illuminating,' or 'inspiring confidence'
are more apt descriptors applying to models than 'valid'"" [Greenberger et al.,
Models in the Policy Process (1976)].

George P. Richardson
Chair of public administration and policy
Rockefeller College of Public Affairs and Policy
University at Albany - SUNY, Albany, NY 12222
Posted by George Richardson <gpr@albany.edu>
posting date Thu, 4 Oct 2007 13:49:11 -0400

_______________________________________________
John Gunkler <jgunkler@sprint
Newbie
Posts: 1
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by John Gunkler <jgunkler@sprint »

Posted by ""John Gunkler"" <jgunkler@sprintmail.com>

Monte writes [numbers added]:

""(1) If the real structure changes, the model of past behavior would
lack the ability to generate the future behavior. (2) It is therefore
hard believe that SD models are theories, unless the structures
of the aspects being modeled are stable.

(3) SD is not effective in dealing with social problems, of which
the dynamics of the system structures are not yet predictable.""

Point (1): Can't argue with this. We build models, in part at least, to
understand why a system behaved the way it did in the past AND to predict
the way it will behave in the future ABSENT ANY FUNDAMENTAL CHANGES IN
STRUCTURE. We're saying, essentially, that ""unless you change something
important to the system's dynamics, here's what you can expect."" Even the
theories of physics have (stated or assumed) limitations -- e.g., Newton's
laws of motion don't apply precisely where there is a very strong
gravitational field; Einstein's General Relativity doesn't apply in the
sub-atomic world; etc. So, in SD models, our ""limitation"" is that the
fundamental causal dynamic structure stays intact.

Point (2): My understanding of what is a scientific theory, from the
philosophy of science I studied, is that it is essentially a prediction of
future behavior of the entity under discussion. How is an SD model (as
described in Point (1)) not a theory, then?

Point (3): I would suggest that the referent of ""the dynamics of the system
structures"" has changed from what it was in Points (1) and (2). In the
previous sentences we are discussing (I think) the fundamental structures
that are responsible for the dynamic behavior of the system. In Point (3),
I suspect, we are now really talking about system behavior. And, in fact,
SD models have a LOT to say about how systems will predictably behave -- how
various structures will interact, how those interactions will affect the
behaviors/outputs of the system as a whole, etc., etc. One of the strengths
of an SD model is that it makes the complex ""dynamics of system structures""
predictable.

Finally, I will admit that the issue of fundamental system structure changes
is very interesting to me. In the current state of SD thinking, I believe
what we say about fundamental system structure changes can be boiled down to
these two things (oversimplifying somewhat):

1. A System Dynamics model helps us understand what ARE the fundamental
causal structures in a system. This, then, will allow us to monitor reality
and look for changes to those fundamental structures, which will then lead
us to change the model and, thus, be able to predict future behavior. It
highlights what is important, and helps us to ignore what isn't --
tremendously useful in the complex situations for which we use SD.

2. A System Dynamics model, by showing us what the fundamental causal
structures are, can help us make deliberate changes in those structures that
will lead to predictable changes in system behavior. A good SD model can
show us where to ""put the lever that will move the world.""

Here's where I agree with Monte. I would hope to see, within the SD method,
a better way to model structural changes. For clarity, let's use Jay
Forrester term ""policies"" to describe a fundamental model structure.
""Policies"" are represented by rate equations. Policies change in the real
world, and there dynamic mechanisms behind (some of) these changes. I wish
there were a better way in SD modeling to show how one policy (rate
equation) can become a different one. We could then have models that
adapted to, or learned from, their environments. While I can think of some
workarounds to do this (e.g., embedding an IF-THEN structure in a rate
equation), I can't think of ways that maintain the transparency that is the
strength of the SD method.

John Gunkler
Posted by ""John Gunkler"" <jgunkler@sprintmail.com>
posting date Thu, 4 Oct 2007 09:50:27 -0400

_______________________________________________
George Richardson <gpr@albany
Member
Posts: 20
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by George Richardson <gpr@albany »

Posted by George Richardson <gpr@albany.edu>

The writer of this query suggests that system dynamics can not be effective
in dealing with social problems because we don't know future social system
structure. It's worse than that, I suppose, because we can't even know
current social system structure for sure. As he says, we can not ""validate""
structure, and we have no test to determine if structure is ""stable"" over
time.

If I understand the posting correctly, the implication appears to be that we
should stop thinking about future social problems because we don't know how
current social structure will evolve. Not only should we not model difficult
current societal problems, we should stop thinking about them in any form,
since the social structures we'd assume in our thoughts can't be ""validated.""

I prefer to think that we have to do the best we can to think about societal
dynamics as they move from the past into the future, rather than refuse to
try to apply the best tools we currently have for thinking about complex
dynamic systems.

It helps our courage to remember that ""No model has ever been or ever will
be thoroughly validated. ... 'Useful,' 'illuminating,' or 'inspiring confidence'
are more apt descriptors applying to models than 'valid'"" [Greenberger et al.,
Models in the Policy Process (1976)].

George P. Richardson
Chair of public administration and policy
Rockefeller College of Public Affairs and Policy
University at Albany - SUNY, Albany, NY 12222
Posted by George Richardson <gpr@albany.edu>
posting date Thu, 4 Oct 2007 13:49:11 -0400

_______________________________________________
""John Gunkler"" <jgunkler@sp
Member
Posts: 20
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by ""John Gunkler"" <jgunkler@sp »

Posted by ""John Gunkler"" <jgunkler@sprintmail.com>

Monte writes [numbers added]:

""(1) If the real structure changes, the model of past behavior would
lack the ability to generate the future behavior. (2) It is therefore
hard believe that SD models are theories, unless the structures
of the aspects being modeled are stable.

(3) SD is not effective in dealing with social problems, of which
the dynamics of the system structures are not yet predictable.""

Point (1): Can't argue with this. We build models, in part at least, to
understand why a system behaved the way it did in the past AND to predict
the way it will behave in the future ABSENT ANY FUNDAMENTAL CHANGES IN
STRUCTURE. We're saying, essentially, that ""unless you change something
important to the system's dynamics, here's what you can expect."" Even the
theories of physics have (stated or assumed) limitations -- e.g., Newton's
laws of motion don't apply precisely where there is a very strong
gravitational field; Einstein's General Relativity doesn't apply in the
sub-atomic world; etc. So, in SD models, our ""limitation"" is that the
fundamental causal dynamic structure stays intact.

Point (2): My understanding of what is a scientific theory, from the
philosophy of science I studied, is that it is essentially a prediction of
future behavior of the entity under discussion. How is an SD model (as
described in Point (1)) not a theory, then?

Point (3): I would suggest that the referent of ""the dynamics of the system
structures"" has changed from what it was in Points (1) and (2). In the
previous sentences we are discussing (I think) the fundamental structures
that are responsible for the dynamic behavior of the system. In Point (3),
I suspect, we are now really talking about system behavior. And, in fact,
SD models have a LOT to say about how systems will predictably behave -- how
various structures will interact, how those interactions will affect the
behaviors/outputs of the system as a whole, etc., etc. One of the strengths
of an SD model is that it makes the complex ""dynamics of system structures""
predictable.

Finally, I will admit that the issue of fundamental system structure changes
is very interesting to me. In the current state of SD thinking, I believe
what we say about fundamental system structure changes can be boiled down to
these two things (oversimplifying somewhat):

1. A System Dynamics model helps us understand what ARE the fundamental
causal structures in a system. This, then, will allow us to monitor reality
and look for changes to those fundamental structures, which will then lead
us to change the model and, thus, be able to predict future behavior. It
highlights what is important, and helps us to ignore what isn't --
tremendously useful in the complex situations for which we use SD.

2. A System Dynamics model, by showing us what the fundamental causal
structures are, can help us make deliberate changes in those structures that
will lead to predictable changes in system behavior. A good SD model can
show us where to ""put the lever that will move the world.""

Here's where I agree with Monte. I would hope to see, within the SD method,
a better way to model structural changes. For clarity, let's use Jay
Forrester term ""policies"" to describe a fundamental model structure.
""Policies"" are represented by rate equations. Policies change in the real
world, and there dynamic mechanisms behind (some of) these changes. I wish
there were a better way in SD modeling to show how one policy (rate
equation) can become a different one. We could then have models that
adapted to, or learned from, their environments. While I can think of some
workarounds to do this (e.g., embedding an IF-THEN structure in a rate
equation), I can't think of ways that maintain the transparency that is the
strength of the SD method.

John Gunkler
Posted by ""John Gunkler"" <jgunkler@sprintmail.com>
posting date Thu, 4 Oct 2007 09:50:27 -0400

_______________________________________________
Jack Homer <jhomer@comcast.ne
Newbie
Posts: 1
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by Jack Homer <jhomer@comcast.ne »

Posted by ""Jack Homer"" <jhomer@comcast.net>

John Gunkler writes:
>I wish there were a better way in SD modeling to show how one policy (rate
>equation) can become a different one. We could then have models that
>adapted to, or learned from, their environments. While I can think of some
>workarounds to do this (e.g., embedding an IF-THEN structure in a rate
>equation), I can't think of ways that maintain the transparency that is the
>strength of the SD method.

I have heard it said before that SD fails to depict structural change, but
I've never understood that. Our models depict behavioral patterns rather
than individual events, and therefore one should not expect to see discrete
instances of structural change in an SD model as one might in an agent-based
model. But that's OK. From the behavioral-pattern perspective, a radical
change in policy(i.e., a structural change), is seen as a shifting in the
weighting among competing priorities, rather than as the appearance of new
priorities out of nowhere. Our behavioral description of this phenomenon is
""shifting loop dominance"". SD is the only social science methodology that
maintains a broad, strategic view (avoiding getting lost in the tactical
details), but still is able to anticipate radical shifts in behavior by
virtue of the interaction of feedback loops and nonlinearities.

Jack Homer
Posted by ""Jack Homer"" <jhomer@comcast.net>
posting date Sat, 6 Oct 2007 01:04:50 -0400

_______________________________________________
Jay Forrester <jforestr@MIT.E
Junior Member
Posts: 12
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by Jay Forrester <jforestr@MIT.E »

Posted by Jay Forrester <jforestr@MIT.EDU>

On Oct 4, 2007, at 6:27 AM, SDMAIL Monte Kietpawpan wrote:
> Structual validation is key to the success of model development.
> However, we have no test to determine whether a given model structure
> is stable over the time horizon of interest.

A model is a hypothesis (a theory) about a structure that one proposes is
causing behavior of interest. The model is a theory in the same sense that
Ohm's Law in electricity is a theory or Einstein's Theory or Relativity.
There are no proofs of the general validity of such theories that are
proposed to represent the real world, one only builds confidence in the
reasonableness and usefulness of such theories within some domain of
applicability. For a system dynamics model, part of the asserted theory
relates to the time dimension and the space dimension of applicability.
By the space dimension, I mean the degree to which a model fits, given
parameters appropriate to the specific situations, all members of the
class of systems to which the system of interest belongs; is the model
generic with broad applicability across systems as well as across time.

> It is therefore
> hard believe that SD models are theories, unless the structures
> of the aspects being modeled are stable.

The duration of stability of structure is determined by how deeply one
chooses to dig into the decision-making processes. Let me illustrate with
an example. In ordering to maintain an inventory, there is a goal as to
the proper amount of inventory. But what is that goal, how is it defined,
and over what time horizon should it be valid? In the short run, the goal
might be to keep 1000 units in inventory, but that would not be responsive
to rising or falling demand for the product. So, one could go further
behind the scenes and say that the inventory should be equal to four weeks
worth of sales and then the inventory would rise and fall with longer-term
average sales. But in the still longer term. why should the inventory be a
fixed number of weeks of sales because that might change with the state of
technology and with the speed and adequacy of communications. What is a
deeper objective? It might be to be able to fill a given fraction, like
95%, of incoming orders from stock. So, the history of order fulfillment
might be invoked to alter the number of weeks of sales to maintain.

Time horizon is important, but it is incorporated as part of the modeling
process and as part of the alleged domain of usefulness of the model.

> SD is not effective in dealing with social problems, of which
> the dynamics of the system structures are not yet predictable.

At a deep enough level, the social structures change slowly, or not at all.
We can still draw relevant insights about human behavior from the writings
of the ancient Greeks. Histories of the rise and fall of civilizations show
a repeating dynamic of growing beyond the capacity of the local natural
environment and then collapsing, the deep human structure of behavior
survives for centuries to produce repeated disasters. The issue here does
not arise from impossibility of modeling to fit a desired time horizon, but
rather depends on the skill of the person doing the modeling and the
correctness of the assertion about the domain of usefulness of a particular
model.

_______________________________________________

Jay W. Forrester
Professor of Management
Sloan School, MIT
Posted by Jay Forrester <jforestr@MIT.EDU>
posting date Fri, 5 Oct 2007 21:47:01 -0400

_______________________________________________
""Jack Homer"" <jhomer@comcas
Member
Posts: 21
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by ""Jack Homer"" <jhomer@comcas »

Posted by ""Jack Homer"" <jhomer@comcast.net>

John Gunkler writes:
>I wish there were a better way in SD modeling to show how one policy (rate
>equation) can become a different one. We could then have models that
>adapted to, or learned from, their environments. While I can think of some
>workarounds to do this (e.g., embedding an IF-THEN structure in a rate
>equation), I can't think of ways that maintain the transparency that is the
>strength of the SD method.

I have heard it said before that SD fails to depict structural change, but
I've never understood that. Our models depict behavioral patterns rather
than individual events, and therefore one should not expect to see discrete
instances of structural change in an SD model as one might in an agent-based
model. But that's OK. From the behavioral-pattern perspective, a radical
change in policy(i.e., a structural change), is seen as a shifting in the
weighting among competing priorities, rather than as the appearance of new
priorities out of nowhere. Our behavioral description of this phenomenon is
""shifting loop dominance"". SD is the only social science methodology that
maintains a broad, strategic view (avoiding getting lost in the tactical
details), but still is able to anticipate radical shifts in behavior by
virtue of the interaction of feedback loops and nonlinearities.

Jack Homer
Posted by ""Jack Homer"" <jhomer@comcast.net>
posting date Sat, 6 Oct 2007 01:04:50 -0400

_______________________________________________
Jay Forrester <jforestr@MIT.E
Junior Member
Posts: 12
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by Jay Forrester <jforestr@MIT.E »

Posted by Jay Forrester <jforestr@MIT.EDU>

On Oct 4, 2007, at 6:27 AM, SDMAIL Monte Kietpawpan wrote:
> Structual validation is key to the success of model development.
> However, we have no test to determine whether a given model structure
> is stable over the time horizon of interest.

A model is a hypothesis (a theory) about a structure that one proposes is
causing behavior of interest. The model is a theory in the same sense that
Ohm's Law in electricity is a theory or Einstein's Theory or Relativity.
There are no proofs of the general validity of such theories that are
proposed to represent the real world, one only builds confidence in the
reasonableness and usefulness of such theories within some domain of
applicability. For a system dynamics model, part of the asserted theory
relates to the time dimension and the space dimension of applicability.
By the space dimension, I mean the degree to which a model fits, given
parameters appropriate to the specific situations, all members of the
class of systems to which the system of interest belongs; is the model
generic with broad applicability across systems as well as across time.

> It is therefore
> hard believe that SD models are theories, unless the structures
> of the aspects being modeled are stable.

The duration of stability of structure is determined by how deeply one
chooses to dig into the decision-making processes. Let me illustrate with
an example. In ordering to maintain an inventory, there is a goal as to
the proper amount of inventory. But what is that goal, how is it defined,
and over what time horizon should it be valid? In the short run, the goal
might be to keep 1000 units in inventory, but that would not be responsive
to rising or falling demand for the product. So, one could go further
behind the scenes and say that the inventory should be equal to four weeks
worth of sales and then the inventory would rise and fall with longer-term
average sales. But in the still longer term. why should the inventory be a
fixed number of weeks of sales because that might change with the state of
technology and with the speed and adequacy of communications. What is a
deeper objective? It might be to be able to fill a given fraction, like
95%, of incoming orders from stock. So, the history of order fulfillment
might be invoked to alter the number of weeks of sales to maintain.

Time horizon is important, but it is incorporated as part of the modeling
process and as part of the alleged domain of usefulness of the model.

> SD is not effective in dealing with social problems, of which
> the dynamics of the system structures are not yet predictable.

At a deep enough level, the social structures change slowly, or not at all.
We can still draw relevant insights about human behavior from the writings
of the ancient Greeks. Histories of the rise and fall of civilizations show
a repeating dynamic of growing beyond the capacity of the local natural
environment and then collapsing, the deep human structure of behavior
survives for centuries to produce repeated disasters. The issue here does
not arise from impossibility of modeling to fit a desired time horizon, but
rather depends on the skill of the person doing the modeling and the
correctness of the assertion about the domain of usefulness of a particular
model.

_______________________________________________

Jay W. Forrester
Professor of Management
Sloan School, MIT
Posted by Jay Forrester <jforestr@MIT.EDU>
posting date Fri, 5 Oct 2007 21:47:01 -0400

_______________________________________________
Alan McLucas <A.McLucas@adfa.
Newbie
Posts: 1
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by Alan McLucas <A.McLucas@adfa. »

Posted by ""Alan McLucas"" <A.McLucas@adfa.edu.au>

I agree that past social structures continue to evolve to become future
social structures, and it is challenging (practically impossible, I
suggest) to frame those structures in a forensic sense with the aim of
build models that we can then expect to validate. However, we can still
learn about these changing structures and use that learning very
effectively to inform judgements about how we might intervene in
problematical situations. Along the way the models we build are
transient objects to inform our learning. The value of those models as
tools for learning is not lessened because they cannot be validated: no
body of social knowledge comparable to the accumulated knowledge of the
physical sciences has been produced - or probably ever will be. So,
validation of our models against some ""truth criterion"" is never going
to be possible. However, I can think of no better way to develop deep
understandings than through building a series of system dynamics models
designed to enable learning, even if all of those models are 'wrong'
insofar as that they can never be fully validated.

We interact with real-world social structures, which present as a flux
of events of ideas unrolling through time: we cannot expect to discover
everything about those social structures all at once (if for no other
reason than they are both complex and capricious). Pragmatically, we
might focus on characterizing those with a view to building better
understandings of them.

Earlier experiences and interactions with those social structures
already would have led us to construct certain mental models of them.
We use these to help us interpret more recent experiences we might have,
as we have them. In so doing, our norms and values are influenced. We
continually analyse experiences and make new interpretations to build or
reframe our models. These models might be either qualitative (our
personal constructs or mental models) or quantitative (system dynamics
models, for example). These models, and related simulations, assist us
in designing our actions (intervention strategies). Even as we
implement our strategies we can expect that social structures will be
changing, but assuming that our monitoring and (management and) control
processes are sensitive to those changes, we should appreciate that we
are now dealing with something which is an evolution of an old
structure. We continually develop an understanding of how changes might
have occurred, formulating and testing our theories as we proceed.

To me this is a strong case for systems thinking and system dynamics
modelling, not an argument against.

Dr Alan McLucas
School of Information Technology and Electrical Engineering,
UNSW@ADFA,
Australian Defence Force Academy,
Northcott Drive,
CAMBPELL ACT 2600
AUSTRALIA
Posted by ""Alan McLucas"" <A.McLucas@adfa.edu.au>
posting date Mon, 8 Oct 2007 16:24:51 +1000

_______________________________________________
""Alan McLucas"" <a.mclucas@a
Junior Member
Posts: 10
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by ""Alan McLucas"" <a.mclucas@a »

Posted by ""Alan McLucas"" <A.McLucas@adfa.edu.au>

I agree that past social structures continue to evolve to become future
social structures, and it is challenging (practically impossible, I
suggest) to frame those structures in a forensic sense with the aim of
build models that we can then expect to validate. However, we can still
learn about these changing structures and use that learning very
effectively to inform judgements about how we might intervene in
problematical situations. Along the way the models we build are
transient objects to inform our learning. The value of those models as
tools for learning is not lessened because they cannot be validated: no
body of social knowledge comparable to the accumulated knowledge of the
physical sciences has been produced - or probably ever will be. So,
validation of our models against some ""truth criterion"" is never going
to be possible. However, I can think of no better way to develop deep
understandings than through building a series of system dynamics models
designed to enable learning, even if all of those models are 'wrong'
insofar as that they can never be fully validated.

We interact with real-world social structures, which present as a flux
of events of ideas unrolling through time: we cannot expect to discover
everything about those social structures all at once (if for no other
reason than they are both complex and capricious). Pragmatically, we
might focus on characterizing those with a view to building better
understandings of them.

Earlier experiences and interactions with those social structures
already would have led us to construct certain mental models of them.
We use these to help us interpret more recent experiences we might have,
as we have them. In so doing, our norms and values are influenced. We
continually analyse experiences and make new interpretations to build or
reframe our models. These models might be either qualitative (our
personal constructs or mental models) or quantitative (system dynamics
models, for example). These models, and related simulations, assist us
in designing our actions (intervention strategies). Even as we
implement our strategies we can expect that social structures will be
changing, but assuming that our monitoring and (management and) control
processes are sensitive to those changes, we should appreciate that we
are now dealing with something which is an evolution of an old
structure. We continually develop an understanding of how changes might
have occurred, formulating and testing our theories as we proceed.

To me this is a strong case for systems thinking and system dynamics
modelling, not an argument against.

Dr Alan McLucas
School of Information Technology and Electrical Engineering,
UNSW@ADFA,
Australian Defence Force Academy,
Northcott Drive,
CAMBPELL ACT 2600
AUSTRALIA
Posted by ""Alan McLucas"" <A.McLucas@adfa.edu.au>
posting date Mon, 8 Oct 2007 16:24:51 +1000

_______________________________________________
ybarlas@boun.edu.tr
Junior Member
Posts: 2
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by ybarlas@boun.edu.tr »

Posted by ybarlas@boun.edu.tr

Quoting SDMAIL Monte Kietpawpan <kietpawpan@yahoo.com>:
> >
> > No existing validity
> > tests serve to check whether the relationships in the real system
> > will not significantly change in the future--within the time horizon.
> > ...

Dear Monte,

Your questions are important. It is true (and essential) that any given system
dynamics model assumes that the corresponding system structure does not
significantly change in the time horizon. Indeed, in testing the validity of a
model, we are testing if we can come up with a simplified, constant
approximation to the real structure (that is for sure changing in reality) that
can reproduce the essential dynamical behavior patterns of the real problem.
Our model, our dynamic hypothesis is in a sense a constant structural
approximation to very messy, changing structures. This is why there is no
'valid' or invalid models; there is a continuum of model validity: terrible,
bad, OK, good, great, superb!...

About your assertion that I quoted above: It is still possible to model and
test problems that involve structures that change significantly in time.
We then build SD models that try to represent the way the structures change.
This 'meta' model will still have a constant structure, but this 'meta' constancy
is now about the rules that are assumed to govern the structural changes in the
real problem. And when we test this type of model, we would be testing your
question quoted above. Not many of us build such models, but it is perfectly
possible within the realm of system dynamics modeling.

thanks and all the mest,

Yaman Barlas
Posted by ybarlas@boun.edu.tr
posting date Tue, 9 Oct 2007 20:18:12 +0300
_______________________________________________
ybarlas@boun.edu.tr Quoting S
Newbie
Posts: 1
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by ybarlas@boun.edu.tr Quoting S »

Posted by ybarlas@boun.edu.tr

Quoting SDMAIL Monte Kietpawpan <kietpawpan@yahoo.com>:
> >
> > No existing validity
> > tests serve to check whether the relationships in the real system
> > will not significantly change in the future--within the time horizon.
> > ...

Dear Monte,

Your questions are important. It is true (and essential) that any given system
dynamics model assumes that the corresponding system structure does not
significantly change in the time horizon. Indeed, in testing the validity of a
model, we are testing if we can come up with a simplified, constant
approximation to the real structure (that is for sure changing in reality) that
can reproduce the essential dynamical behavior patterns of the real problem.
Our model, our dynamic hypothesis is in a sense a constant structural
approximation to very messy, changing structures. This is why there is no
'valid' or invalid models; there is a continuum of model validity: terrible,
bad, OK, good, great, superb!...

About your assertion that I quoted above: It is still possible to model and
test problems that involve structures that change significantly in time.
We then build SD models that try to represent the way the structures change.
This 'meta' model will still have a constant structure, but this 'meta' constancy
is now about the rules that are assumed to govern the structural changes in the
real problem. And when we test this type of model, we would be testing your
question quoted above. Not many of us build such models, but it is perfectly
possible within the realm of system dynamics modeling.

thanks and all the mest,

Yaman Barlas
Posted by ybarlas@boun.edu.tr
posting date Tue, 9 Oct 2007 20:18:12 +0300
_______________________________________________
Tom Fiddaman <tom@ventanasyst
Member
Posts: 25
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by Tom Fiddaman <tom@ventanasyst »

Posted by Tom Fiddaman <tom@ventanasystems.com>


>No existing validity tests serve to check whether the relationships in the real system
>will not significantly change in the future--within the time horizon.

I don't believe that this is strictly true.

For example, as a matter of habit I include a first-order negative feedback constraint
on inventories, which guarantees that the outflow (sales) is zero when the contents of
the inventory is zero. I don't recall ever working with data that demonstrated that
constraint in action, but its existence is common sense. Extreme conditions tests and a
variety of other sources of knowledge can reveal similar latent feedback loops that
might become active at some point in the future.

Perhaps my example is most relevant to physical systems, where conservation laws and so
forth somewhat constrain the possible structural changes. Still, even in models of
social systems, it is possible to speculate about feedbacks that might emerge. I could
imagine testing a model by randomly adding terms to every rate equation that created
arbitrary perturbations of the system structure, in order to discover which, if any,
might be interesting or important.

We may now be witnessing the effects of climate feedbacks that represent changes in
loop dominance, e.g. rapid progression of ice-albedo effects in polar ice. The outcome
of those feedbacks is surprising in the context of global models, which are constructed
with a bias toward avoiding making concrete statements about ill-parameterized effects.
Yet the presence of novel feedbacks as the system is pushed outside its historical
envelope is hardly surprising. The particulars of the social response to climate news
is probably more unpredictable, but we could similarly expect the emergence of novel
behavior.

The point is well taken that some changes in system structure cannot be anticipated. All
theories, whether formalized as mathematical models or not, are subject to such
surprises. This should not stop us from modeling social systems. The real trouble with
surprise arises from a forecasting approach to problem solving: the fallacy that ""if I
know what will happen I'll know what to do."" If one instead uses models to build robust
feedback controls that work over a wide range of conditions, surprises from latent
structure are not so problematic.

I think there are social phenomena that are difficult to represent in traditional
aggregate SD models. Agent based models are a useful alternative for exploring their
dynamics, but I don't think their track record for prediction or policy making is
conspicuously better, nor are they somehow theories when other formalisms are not.
I view the two approaches as complementary and frequently overlapping.

Tom

****************************************************
Tom Fiddaman
Ventana Systems, Inc.
Posted by Tom Fiddaman <tom@ventanasystems.com>
posting date Fri, 12 Oct 2007 21:11:32 -0600
_______________________________________________
Tom Fiddaman <tom@ventanasyst
Member
Posts: 25
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by Tom Fiddaman <tom@ventanasyst »

Posted by Tom Fiddaman <tom@ventanasystems.com>


>No existing validity tests serve to check whether the relationships in the real system
>will not significantly change in the future--within the time horizon.

I don't believe that this is strictly true.

For example, as a matter of habit I include a first-order negative feedback constraint
on inventories, which guarantees that the outflow (sales) is zero when the contents of
the inventory is zero. I don't recall ever working with data that demonstrated that
constraint in action, but its existence is common sense. Extreme conditions tests and a
variety of other sources of knowledge can reveal similar latent feedback loops that
might become active at some point in the future.

Perhaps my example is most relevant to physical systems, where conservation laws and so
forth somewhat constrain the possible structural changes. Still, even in models of
social systems, it is possible to speculate about feedbacks that might emerge. I could
imagine testing a model by randomly adding terms to every rate equation that created
arbitrary perturbations of the system structure, in order to discover which, if any,
might be interesting or important.

We may now be witnessing the effects of climate feedbacks that represent changes in
loop dominance, e.g. rapid progression of ice-albedo effects in polar ice. The outcome
of those feedbacks is surprising in the context of global models, which are constructed
with a bias toward avoiding making concrete statements about ill-parameterized effects.
Yet the presence of novel feedbacks as the system is pushed outside its historical
envelope is hardly surprising. The particulars of the social response to climate news
is probably more unpredictable, but we could similarly expect the emergence of novel
behavior.

The point is well taken that some changes in system structure cannot be anticipated. All
theories, whether formalized as mathematical models or not, are subject to such
surprises. This should not stop us from modeling social systems. The real trouble with
surprise arises from a forecasting approach to problem solving: the fallacy that ""if I
know what will happen I'll know what to do."" If one instead uses models to build robust
feedback controls that work over a wide range of conditions, surprises from latent
structure are not so problematic.

I think there are social phenomena that are difficult to represent in traditional
aggregate SD models. Agent based models are a useful alternative for exploring their
dynamics, but I don't think their track record for prediction or policy making is
conspicuously better, nor are they somehow theories when other formalisms are not.
I view the two approaches as complementary and frequently overlapping.

Tom

****************************************************
Tom Fiddaman
Ventana Systems, Inc.
Posted by Tom Fiddaman <tom@ventanasystems.com>
posting date Fri, 12 Oct 2007 21:11:32 -0600
_______________________________________________
ybarlas@boun.edu.tr
Junior Member
Posts: 2
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by ybarlas@boun.edu.tr »

Posted by <ybarlas@boun.edu.tr>

Hi again,

I received a couple of emails on what I wrote on this topic (that one could
model meta-dynamics of structural changes. My replies are extensions and
illustrations of George's (Richardson) example on modeling structural changes
using exponential averages.

The following more practical clues could be perhaps useful to some
folks interested in modeling such phenomena:

The key method is to use a 'potential' model structure that is much more
comprehensive/larger than the one needed in any time period. Then one could use
exponentially weighted averages that would gradually phase out some structures
and introduce new ones (with weights that are dynamically determined in time,
some approaching zero while others emerging from zero ). This, of course, if
we assume that structural changes are not sudden and revolutionary, but rather
evolutionary. In case of sudden structural changes that occur as a result of
critical accumulations, we can use IF THEN ELSE structures.

Examples of such structures exist in models dealing with goal formation. A
recent and comprehensive version(s) can be found in:

- Barlas Y. and Hakan Ya?arcan. A Comprehensive Model Of Goal Dynamics In
Organizations: Setting, Evaluation And Revision. To appear in Complex Decision
Making: Theory and Practice, Springer, 2007.

- Barlas Y. and Hakan Ya?arcan. Goal setting, evaluation, learning and revision:
A dynamic Modeling approach. Evaluation and Program Planning. 29(1). 2006. pp
79-87.

all the best,
Yaman Barlas
---------------------------------------------------------------------------
Yaman Barlas, Ph.D.
Chair, Industrial Engineering Dept.
Bogazici University,
34342 Bebek, Istanbul, TURKEY
Posted by ybarlas@boun.edu.tr
posting date Mon, 15 Oct 2007 11:26:40 +0300
_______________________________________________
<ybarlas@boun.edu.tr>
Junior Member
Posts: 3
Joined: Fri Mar 29, 2002 3:39 am

QUERY Structural changes and validity

Post by <ybarlas@boun.edu.tr> »

Posted by <ybarlas@boun.edu.tr>

Hi again,

I received a couple of emails on what I wrote on this topic (that one could
model meta-dynamics of structural changes. My replies are extensions and
illustrations of George's (Richardson) example on modeling structural changes
using exponential averages.

The following more practical clues could be perhaps useful to some
folks interested in modeling such phenomena:

The key method is to use a 'potential' model structure that is much more
comprehensive/larger than the one needed in any time period. Then one could use
exponentially weighted averages that would gradually phase out some structures
and introduce new ones (with weights that are dynamically determined in time,
some approaching zero while others emerging from zero ). This, of course, if
we assume that structural changes are not sudden and revolutionary, but rather
evolutionary. In case of sudden structural changes that occur as a result of
critical accumulations, we can use IF THEN ELSE structures.

Examples of such structures exist in models dealing with goal formation. A
recent and comprehensive version(s) can be found in:

- Barlas Y. and Hakan Ya?arcan. A Comprehensive Model Of Goal Dynamics In
Organizations: Setting, Evaluation And Revision. To appear in Complex Decision
Making: Theory and Practice, Springer, 2007.

- Barlas Y. and Hakan Ya?arcan. Goal setting, evaluation, learning and revision:
A dynamic Modeling approach. Evaluation and Program Planning. 29(1). 2006. pp
79-87.

all the best,
Yaman Barlas
---------------------------------------------------------------------------
Yaman Barlas, Ph.D.
Chair, Industrial Engineering Dept.
Bogazici University,
34342 Bebek, Istanbul, TURKEY
Posted by ybarlas@boun.edu.tr
posting date Mon, 15 Oct 2007 11:26:40 +0300
_______________________________________________
Locked