System Dynamics vs. Discrete Event Simulation, Expert System

[Niall Palfreyman]

Jim Hines wrote:

> The most important difference is the **purpose** of the modeling. The
> discrete event folks are trying to find how many resources you need ...
> The SD folks are trying to figure out how to change decision making ...
> so that inventories, sales, production, etc. wont oscillate or show
> some other undesirable pattern of behavior.

Brian Dangerfield schrieb:

> You are going over old ground...

I disagree, Brian. I _need_ discussion of the deeper differences and
similarities between SD and other methodologies to understand better
whats going on. Jims contribution emphasises the distinction of
resources vs. behaviour, and this distinction is a new one to me. I find
it reminiscent of the distinction between the more data-centred
entity-relationship modelling and the more behaviour-centred
object-oriented modelling. Is there any similarity between these two
juxtapositions? (Im not at all familiar with discrete-event modelling,
although it sounds similar to frame-based modelling.)

Niall Palfreyman.
From: Niall Palfreyman <niall.palfreyman@fh-weihenstephan.de>
n:Palfreyman;Niall
x-mozilla-html:FALSE
org:Weihenstephan University of Applied Sciences;Bioinformatics
adr:;;Am Hofgarten 10;Freising;;85350;Germany
version:2.1
email;internet:Niall.Palfreyman@fh-weihenstephan.de
title:Dean of Academic Studies
fn:Prof. Dr. Niall Palfreyman

["geoff coyle"]

Its also easy to embed discrete events within an SD model (where the
purpose of the model calls for it). I can provide reference if anyone wants
them.

Regards,

Geoff
From: "geoff coyle" <geoff.coyle@btinternet.com>

Professor Geoff Coyle
Consultant in System Dynamics and Strategic Analysis,
Telephone 44 (0) 1793 782817
Fax 44 (0) 1793 783188

[Joel Rahn]

Not to sow more confusion but it is possible to develop a discrete-event
model of an SD model. A Master`s student of mine and I did so for a small
Commodity Cycle model about 10 years ago. The time behaviours don`t look very
similar but the autocorrelation structures are quite similar. On the other
hand, many SD models are developed by following the path of a typical item
through the physical system structure as one does when developing a
discrete-event process-style stochastic simulation model.


R. Joel Rahn
From: Joel Rahn <rjrahn@videotron.ca>