Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Found Sci (2008) 13:113125 DOI 10.1007/s10699-008-9121-3

Summary: Found Sci (2008) 13:113125
DOI 10.1007/s10699-008-9121-3
A Multiagent Approach to Modelling Complex
Francesco Amigoni Viola Schiaffonati
Published online: 22 May 2008
Springer Science+Business Media B.V. 2008
Abstract Designing models of complex phenomena is a difficult task in engineering that
can be tackled by composing a number of partial models to produce a global model of the
phenomena. We propose to embed the partial models in software agents and to implement
their composition as a cooperative negotiation between the agents. The resulting multiagent
system provides a global model of a phenomenon. We applied this approach in modelling two
complex physiological processes: the heart rate regulation and the glucose-insulin metab-
olism. Beyond the effectiveness demonstrated in these two applications, the idea of using
models associated to software agents to give reason of complex phenomena is in accordance
with current tendencies in epistemology, where it is evident an increasing use of computa-
tional models for scientific explanation and analysis. Therefore, our approach has not only
a practical, but also a theoretical significance: agents embedding models are a technology
suitable both to representing and to investigating reality.
Keywords Computational model Multiagent system


Source: Amigoni, Francesco - Dipartimento di Elettronica e Informazione, Politecnico di Milano


Collections: Computer Technologies and Information Sciences