Estimating global stress environment by observing local behavior in distributed multiagent systems

Seokcheon Lee, Soundar Kumara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

A multiagent system can be considered survivable if it adapts itself to varying stresses without considerable performance degradation. Such an adaptivity comprises of identifying the behavior of the agents in a society, relating them to stress situations, and then invoking control rules. This problem is a hard one, especially in distributed multiagent systems wherein the agent behaviors tend to be nonlinear and dynamic. In this paper, we study a supply chain planning system implemented in COUGAAR (Cognitive Agent Architecture) and develop a methodology for identifying the behavior of agents through their behavioral parameters, and relating those parameters to stress situations. One important aspect of our approach is that we identify the stress situations of agents in the society by observing local behavior of one representative agent. This approach is motivated by the fact that a local time series can have the information of the dynamics of the entire system in deterministic dynamical systems. We validate our approach empirically through identifying the stress situations using k-nearest neighbor algorithm based on the behavioral parameters.

Original languageEnglish (US)
Title of host publicationProceedings of the 2005 IEEE Conference on Automation Science and Engineering, IEEE-CASE 2005
Pages215-219
Number of pages5
DOIs
StatePublished - Dec 1 2005
Event2005 IEEE Conference on Automation Science and Engineering, IEEE-CASE 2005 - Edmonton, Canada
Duration: Aug 1 2005Aug 2 2005

Publication series

NameProceedings of the 2005 IEEE Conference on Automation Science and Engineering, IEEE-CASE 2005
Volume2005

Other

Other2005 IEEE Conference on Automation Science and Engineering, IEEE-CASE 2005
CountryCanada
CityEdmonton
Period8/1/058/2/05

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Lee, S., & Kumara, S. (2005). Estimating global stress environment by observing local behavior in distributed multiagent systems. In Proceedings of the 2005 IEEE Conference on Automation Science and Engineering, IEEE-CASE 2005 (pp. 215-219). [1506771] (Proceedings of the 2005 IEEE Conference on Automation Science and Engineering, IEEE-CASE 2005; Vol. 2005). https://doi.org/10.1109/COASE.2005.1506771