Methods to Observe the Clustering of Agents Within a Multi-Agent System

  • Authors:
  • Raymond J. Staron;Pavel Tichý;Radek Šindelář;Francisco P. Maturana

  • Affiliations:
  • Rockwell Automation, 1 Allen-Bradley Drive, Mayfield Hts., OH 44124-6118, USA;Rockwell Automation s.r.o., Pekarska 10a, 15500 Prague 5, Czech Republic;Rockwell Automation s.r.o., Pekarska 10a, 15500 Prague 5, Czech Republic;Rockwell Automation, 1 Allen-Bradley Drive, Mayfield Hts., OH 44124-6118, USA

  • Venue:
  • HoloMAS '07 Proceedings of the 3rd international conference on Industrial Applications of Holonic and Multi-Agent Systems: Holonic and Multi-Agent Systems for Manufacturing
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Cohesion and coupling are standard concepts in computer science. As in classic software, multi-agent systems (MASs) strive for high cohesion and low coupling. But MASs, by definition, do have some coupling, i.e., the agents seek out and discover other agents, and negotiate and cooperate with them, to accomplish some overall goal. Moreover, as different events occur, and different agents and different agent behaviors are called upon, the coupling values for the system change. This paper proposes a methodology for viewing these changing coupling relationships and clustering behaviors. In addition, the methodology is described as implemented as part of a standard agent monitoring tool.