Analytical equilibrium; presentations

[Jim Thompson <73424.1506@compuse]

1) About two years ago, I got tired of headaches caused by applying algebra to
initial level values for models or model fragments that I wished to test by
starting in equilibrium. I bought one of the symbolic processor programs, Maple
V, and found that I could reduce significantly the time spent on this chore. I
wont describe all the steps I use. Briefly, I equate the level inflows to the
level outflows and "isolate" the level on the left of the equal sign.

This procedure seems to work well, but now Im concerned. Is this short-cut
causing me to make modeling errors that are obscured by the tool? Do others use
symbolic processors (Maple V or Mathematica) to the same end?

2) I havent had much time to invest using Maple V or Mathematica to investigate
their other capabilities. However, I have noted that their graph output is
quite versatile and seems more flexible (configurable) than Vensim or STELLA. I
spend much more time on communicating the results of modeling than on modeling,
and presentation is a very important part of the communication process. I would
be interested to learn of your experiences linking modeling software with
general purpose math software products or other presentation aids.

Jim Thompson
Gemini Consulting, Inc.
55 Reservoir Road
Farmington, CT 06032
phone 860/676-8152
fax 860/676-8052
e-mail 73424.1506@compuserve.com

["588458@KP40"]

To: SYSTEMD3--EKSMTP

*** Reply to note of 02/17/96 00:36
From: William Bonnell
Mgt.Services Div., Eastman Kodak Company
Bldg. 56-4, KP, Rochester, NY 14652-4530
Subject: Analytical equilibrium; presentations
I used a package called DERIVE to do the same thing. The analytical
expression for the equilibrium initialization was over 700 lines of 80 column
algebra. I havent done it again but I think its a good idea.

___________________________
Bill Bonnell (716) 477-7874
bonnell@kodak.com
---------------------------