A framework for performance control of distributed autonomous agents

Nathan Gnanasambandam; Seokcheon Lee; Soundar R.T. Kumara; Natarajan Gautam

A framework for performance control of distributed autonomous agents

Nathan Gnanasambandam, Seokcheon Lee, Soundar R.T. Kumara, Natarajan Gautam

Research output: Contribution to conference › Paper › peer-review

Abstract

We propose an autonomous and scalable queueing theory-based methodology to control the performance of a hierarchical network of distributed agents. Multi-agent systems (MAS) such as supply chains functioning in highly dynamic environments need to achieve maximum overall utility during operation. Hence, the objective of the control framework is to identify the trade-offs between quality and performance and adaptively choose the operational settings to posture the MAS for better utility. By formulating the MAS as an open queueing network with multiple classes of traffic we evaluate the performance and subsequently the utility, from which we identify the control alternative for a localized, multi-tier zone.

Original language	English (US)
State	Published - 2005
Externally published	Yes
Event	IIE Annual Conference and Exposition 2005 - Atlanta, GA, United States Duration: May 14 2005 → May 18 2005

Conference

Conference	IIE Annual Conference and Exposition 2005
Country/Territory	United States
City	Atlanta, GA
Period	5/14/05 → 5/18/05

Keywords

Multi-Agent Systems
Performance control
Queueing Network

ASJC Scopus subject areas

General Engineering

Cite this

@conference{6cac2a1a94484fceb61ce3b1db3e1e23,

title = "A framework for performance control of distributed autonomous agents",

abstract = "We propose an autonomous and scalable queueing theory-based methodology to control the performance of a hierarchical network of distributed agents. Multi-agent systems (MAS) such as supply chains functioning in highly dynamic environments need to achieve maximum overall utility during operation. Hence, the objective of the control framework is to identify the trade-offs between quality and performance and adaptively choose the operational settings to posture the MAS for better utility. By formulating the MAS as an open queueing network with multiple classes of traffic we evaluate the performance and subsequently the utility, from which we identify the control alternative for a localized, multi-tier zone.",

keywords = "Multi-Agent Systems, Performance control, Queueing Network",

author = "Nathan Gnanasambandam and Seokcheon Lee and Kumara, {Soundar R.T.} and Natarajan Gautam",

year = "2005",

language = "English (US)",

note = "IIE Annual Conference and Exposition 2005 ; Conference date: 14-05-2005 Through 18-05-2005",

}

TY - CONF

T1 - A framework for performance control of distributed autonomous agents

AU - Gnanasambandam, Nathan

AU - Lee, Seokcheon

AU - Kumara, Soundar R.T.

AU - Gautam, Natarajan

PY - 2005

Y1 - 2005

N2 - We propose an autonomous and scalable queueing theory-based methodology to control the performance of a hierarchical network of distributed agents. Multi-agent systems (MAS) such as supply chains functioning in highly dynamic environments need to achieve maximum overall utility during operation. Hence, the objective of the control framework is to identify the trade-offs between quality and performance and adaptively choose the operational settings to posture the MAS for better utility. By formulating the MAS as an open queueing network with multiple classes of traffic we evaluate the performance and subsequently the utility, from which we identify the control alternative for a localized, multi-tier zone.

AB - We propose an autonomous and scalable queueing theory-based methodology to control the performance of a hierarchical network of distributed agents. Multi-agent systems (MAS) such as supply chains functioning in highly dynamic environments need to achieve maximum overall utility during operation. Hence, the objective of the control framework is to identify the trade-offs between quality and performance and adaptively choose the operational settings to posture the MAS for better utility. By formulating the MAS as an open queueing network with multiple classes of traffic we evaluate the performance and subsequently the utility, from which we identify the control alternative for a localized, multi-tier zone.

KW - Multi-Agent Systems

KW - Performance control

KW - Queueing Network

UR - http://www.scopus.com/inward/record.url?scp=33749247617&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33749247617&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:33749247617

T2 - IIE Annual Conference and Exposition 2005

Y2 - 14 May 2005 through 18 May 2005

ER -

A framework for performance control of distributed autonomous agents

Abstract

Conference

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this