Spaghetti: Hierarchy the cure?

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
"William H. Cutler" <72734.3452@
Junior Member
Posts: 11
Joined: Fri Mar 29, 2002 3:39 am

Spaghetti: Hierarchy the cure?

Post by "William H. Cutler" <72734.3452@ »

Paul Kucera writes, in reply to my suggestion to use hierarchy as an antidote to
spaghetti-itis:

>Good SD models are almost often built top down as was
>suggested, but using the stock flow language can often be more helpful in
>getting to meangingful structure than can applying some exogenous structural
>framework like the one described.

The purpose of the modeling is insight. Insight can be lost in the spaghetti,
or can be lost in the coarse-graining of some hierarchical structure. I cant
provide any guaranteed algorithm for getting it right. The proper balance
between too much detail and too coarse-grained depends on the skill of the
modeller. However, there are tools and heuristics which can help. Looking at
general properties of the system such as centralized/distributed,
horizontal/vertical, tightly/loosly coupled, can provide clues.

As I mentioned in my earlier message, the N-squared chart is a powerful tool for
discovering how to chunk the system to make it comprehensible without loosing
any of the richness of the structure. The entities entered into the N-square
can be the functional modules of a stock and flow if you like. You can make it
as big as needed, without limit, to encompass the system you are describing.

I think that hierarchy is a useful tool in most cases. Very few systems are so
flat that they dont exhibit a hierarchical structure.

I dont think Im >applying some exogenous structural framework< by proposing
such tools and approaches, any more than stock and flow is an exogenous
structural framework. All of these are abstractions which help our human
pattern recognition capability get a grip on the problem. If the pattern gets
lost in the fine-graining, well never see it. The skill and art is to let the
system tell us what its structure is.

Im still looking for an answer to my general question: is the network/hierarchy
structure necessary and sufficient to describe any and all systems, or is there
some fundamentally different representation which will also do the job?

Bill Cutler
72734.3452@compuserve.com
Locked