** Next:** 2. Model Representation Transformation
** Up:** 2. The Proposal
** Previous:** 2. The Proposal

It is investigated how to
systematically derive models of different complexity. These may be
simplified models in the same formalism [50] but also more abstract
models in a different
representation [44]. This methodology
can be applied, e.g.,
- to perform optimization with increasingly
complex models, which may be more likely to find a global
optimum [20],
- to allow vendor models (destined to replace electronic data sheets)
to be combined into one extensive
and complex
*base model* that can be used in a reduced form for
the different design and operation tasks
(e.g., control design, performance
assessment, and model-based diagnosis) [38],
- to design
intelligent numerical solvers that adapt the complexity of the
model to the efficiency
requirements (e.g., real-time simulation constraints) [30],
- to support reactive learning environments
(so called
*microworlds*) by
increasingly adding detail to the world model [42], and
- to infer the required level of detail of model parts in different
representations to ensure consistency of analysis results of the
overall combined model against given criteria.

Note that in general it may
be possible to automatically add model detail as well as to automatically
reduce complexity of a base model [9].

** Next:** 2. Model Representation Transformation
** Up:** 2. The Proposal
** Previous:** 2. The Proposal
*Pieter Mosterman ER*

*2001-06-19*