Improved performance of agent based placement cell system: a performance efficient role clustering technique

  • Authors:

  • Soumya Suravita;Prabhat Ranjan;R. K. Singh;A. K. Misra

  • Affiliations:

  • Department of Computer Science & Engineering, Motilal Nehru National Institute of Technology, Allahabad, Uttar Pradesh, India;Department of Computer Science & Engineering, Motilal Nehru National Institute of Technology, Allahabad, Uttar Pradesh, India;Electrical Engineering Departments, Motilal Nehru National Institute of Technology, Allahabad, Uttar Pradesh, India;Department of Computer Science & Engineering, Motilal Nehru National Institute of Technology, Allahabad, Uttar Pradesh, India

  • Venue:
  • ICCOMP'06 Proceedings of the 10th WSEAS international conference on Computers
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper provides three criterias of clustering within a placement cell multi-agent system. Our goal is to group roles with similar objectives. Optimize the system performance by minimizing the overall interaction, data transmission and competition of shared resource between roles/agents. This paper presents a novel systematic approach to optimize the system performance by exploiting the relationships and dependencies among roles as well as clustering of roles and mapping criteria between roles to agents. The proposed clustering algorithm partition the overall system roles/agents into several clusters. Optimal cluster size can be obtained by user-defined performance parameter (η). The performance of the agent-based system, enhanced with our algorithm are investigated via implementation of placement cell case study. The results indicate that our proposed algorithm enhances the system performance if intersections of the cluster are minimal.