Team Formation Strategies in a Dynamic Large-Scale Environment

  • Authors:
  • Chris L. Jones;K. Suzanne Barber

  • Affiliations:
  • Laboratory for Intelligent Processes and Systems, The University of Texas at Austin, Austin 78712-0240;Laboratory for Intelligent Processes and Systems, The University of Texas at Austin, Austin 78712-0240

  • Venue:
  • Massively Multi-Agent Technology
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

In open markets and within business and government organizations, fully autonomous agents may form teams to work on large, multifaceted problems. Factors such as uncertain information, bounded rationality and environmental dynamicism can lead to sudden, unforeseen changes in both solution requirements and team participation. Accordingly, this paper proposes and examines strategies for team formation strategies in a large-scale, dynamic environment. Strategies control how agents select problems to work on and partners to work with. The paper includes an experimental evaluation of the relative utility of each strategy in an increasingly dynamic environment, and concludes that a strategy which combines greedy job selection with adaptive team selection performs best in highly dynamic environments. Alternatively, greedy job selection combined with selecting smaller teams performs best in environments with little to no dynamicism.