Database Reference
In-Depth Information
vaccine only add to the overall maladies. Randomness exists in all living systems.
For example, the effect that air temperature has on the birth rate of mosquitoes car-
rying West Nile virus can be dramatic. However, trying to predict the critical air
tomorrow or over the next two weeks, a long time in the world of the insect, is best
modeled using historic temperatures with an added random increment. Then one
can examine the range of possible outcomes due to this randomness.
Not only is randomness something that one must deal with in models but it is
the source, for good or bad, of the new. Treating the need to consider randomness
in models as necessary when one wanted to represent the variations of say weather,
or temperature - the kinds of variables for which one could not and need not pro-
vide explicit models but whose behavior may appropriately be described by varia-
tion around some mean values. But randomness has a more important aspect: The
epistemologist Gregory Bateson has remarked that we could not have music and
the creation of novel forms unless we had a background of noise, of uncommitted
potential in randomness and disorder that awaited selection in the ordering of the
creative act. “All that is not information, not redundancy, not form and not restraints
- is noise, the only possible source of new patterns.”
1.5 Modeling in STELLA
STELLA
2
was chosen as the computer language for this topic on Dynamic
Modeling of Pestilence because it is a powerful, yet easy-to-learn tool. Readers
are expected to familiarize themselves with the many features of the program. Some
introductory material is provided in the appendix. Careful reading of the Help File
that accompanies the program is advised. Experiment with the STELLA software
and become thoroughly familiar with it.
To explore modeling with STELLA, we will develop a basic model of the dy-
namics of a contagious disease in a human population. Assume that initially, only
10 people are sick and that the contagion rate is 5 percent per day; that is, each day,
5 people become sick for every 100 sick people. For simplicity, assume also that
none of the sick people die. How many sick people will we have after 80 days?
In building the model, utilize all four of the graphical “tools” for programming
in STELLA. The appendix has a “Quick Help Guide” to the software. The appendix
also describes how to install the STELLA software and models of the topic. Follow
these instructions. Then, double-click on the STELLA icon to open it.
On opening STELLA, you will be faced with the “High-Level Mapping Layer,”
which is provided to help you develop user interfaces for an existing model as you
become more experienced. For now, go to the “Diagram Layer”—that layer in which
we actually develop an executable model—and click on the downward-pointing ar-
row in the upper left-hand corner of the frame (Figure 1.2):
2
All models in this topic were made with STELLA complete version 9.
