references

[Bell GA]

Dear system dynamics practitioners

Can anyone direct me to some references on the following:

1) A discussion on first and third generation system dynamics modelling

2) An investigation of cost estimation accuracy in the research and
development field of study i.e. comparisons between system dynamics,
regression models and Nordens model.

This will help me with my thesis titled "A systems approach to software
project cost explanation and estimation"

Regards

Gary Bell

The City University
School of Informatics
Business Computing Department
Northampton Square
London EC1V OHB.
Tel: +44 171 477 8000
Fax: +44 171 477 8586
Email: sc308@city.ac.uk

[George Richardson]

Gary Bell asked for references to "first and third generation system
dynamics modeling."

The notion of such generations is potentially misleading. Forresters
original work in Industrial Dynamics and Robertss work in research and
development are still models of great modeling in the field. I would say
there are no "generational" distinctions, other than software, in the high
quality work in the field.

There have been some dramatic changes in software, which have enabled
changes in practice (such as group model building in real time and the
use of serious games as learning environments). But the underlying
modeling still looks the same and should be evaluated by the same
strict criteria that have been applied for decades in the field.

There are some changes in what modelers think they are modeling: in the
1992 interview in the McKinsey Quarterly, Forrester notes a change from
"modeling a process" to modeling to help people think about a complex
system (I dont have the exact quote at hand).

And there was a lovely speech by Ed Roberts at the 1981 conference in
Renssalaerville, New York, in which he distinquished generations or waves
of modelers -- the changing people dominating the field over time. But
not necessarily generations of standards or modes of practice.

The notion of "generations" suggests something more than a continuous
improvement process in the normal science of our field. It suggests that
what is done now is in the most up-to-date generation and therefore
better than in previous generations. I doubt that is true, and I worry
that it means that newcomers will not go back to learn from the great
work in our past.

...George


-----------------------------------------------------------------------------
George P. Richardson G.P.Richardson@Albany.edu
Rockefeller College of Public Affairs and Policy Phone: 518-442-3859
University at Albany - SUNY, Albany, NY 12222 Fax: 518-442-3398
-----------------------------------------------------------------------------

[saeed@ait.ac.th (Prof Khalid Sae]

I like George Richardsons response to Gary Bells reference to
"generations." However, to infer that there is no difference in the
modeling practices of today and the past spells stagnation and an absence
of evolution in the field, which is not true.

Perhaps we do have several generations of system dynamists in a
chronological sense, who have contributed to the evolution of the field,
although this has not created segments in the field itself since these
generations have learnt from one another. Also, the field is still young
and has inadequate record of its evolution, hence one might need to look at
all significant contributions, both old and new to get to know the current
practice intimately. We do need, however, to continue to strive for
building a comprehensive record of the practice. George has contributed
generously to building such a record both through his text book and his
book on history of feedback. These books are able to address many questions
both about the state of the practice and its history. We must build further
on this work before it becomes generation specific.

Khalid
saeed@ait.ac.th

Professor Khalid Saeed
Infrastructure Planning & Management
School of Civil Engineering
ASIAN INSTITUTE OF TECHNOLOGY
G.P.O. Box 2754, Bangkok, THAILAND
phones: (66-2)524-5681, (66-2)524-5785; fax: (66-2)524-5776

["Teiling, Bernard"]

I also liked the clarity of Richardsons answer to Gary Bells request.
However, I feel quite confused by the implicit references to the
"generations" or "waves" in modeling techniques, as it transpires from this
exchange of messages.
Strictly speaking about software generations, I thought we were already at
number 4 (or 5 with some authors), and it would correspond more or less to
object oriented programming environment.
Can anybody give me a simple historical perspective on the "modeling
generations" (or waves, if needed to avoid confusion with the evolution of
computers systems)?
Bernard Teiling
usgle000.teilinb@wcsmvs.infonet.com
----------