Scalability of Multi-Agent Systems - Proposal for a Dissertation

Christian Gerber

DFKI, DFKI Technical Memos (TM), Vol. 97-02, 1997.


This work proposes a generic approach for achieving scalability of multi-agent-systems (MAS). The key to reach that goal is to introduce a self-organization mechanism allowing systems to configure themselves to any application scale and nature. The work outlines such a mechanism, which can be introduced to any multi-agent system. Focus is put on two systems, InteRRaP and MECCA, and on two applications, the transportation domain, realized using InteRRaP agents, and a traffic telematics application modelled with MECCA agents. All systems and applications are briefly characterized in this work and relations to other fields of research are pointed out.

TM-97-02.pdf (pdf, 406 KB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence