Computer-assisted Modeling of Ill-defined Systems

François E. Cellier
Institute of Computational Science
ETH Zürich
CH-8092 Zürich
Switzerland
Email: FCellier@Inf.ETHZ.CH
Andreas Fischlin
Terrestrial Systems Ecology, CHN E35.1
ETH Zürich
CH-8092 Zürich
Switzerland
Email: Andreas.Fischlin@Env.ETHZ.CH

Abstract

This paper shows that the highly sophisticated simulation software as it is available of today is not too well suited for the analysis of so-called ill-defined systems. In a first section of this paper, we shall identify the difficulties inherent in ill-defined system modeling. The second part then deals with a discussion of possible adaptations and enhancements of simulation software for improving its modeling capabilities. The resulting enhanced simulation languages, which may be more appropriately called "model-oriented" languages, are expected to be much better suited for the derivation of models for ill-defined systems than the currently available simulation software.

For the illustration of our statements, we shall cite examples from biological modeling (larch bud moth), from social modeling (Forrester's WORLD1 model), as well as from psychological modeling (conformity of individuals with the judgement of groups).

Interested in reading the full paper? (13 pages, 1,317,368 bytes, pdf)

Homepage