A Decision Support Algorithm for Ontology-Based Negotiation Agents within Virtual Enterprises

This paper describes an application of a distributed knowledge-based multi-agent architecture to support automated negotiation between and within virtual enterprises (VE). Considering constraints of traditional negotiation approaches (predetermined protocols as well as hard-coded and non-adaptive st...

Full description

Saved in:
Bibliographic Details
Main Authors: Koppensteiner, G., Merdan, M., Lepuschitz, W., Reinprecht, C., Riemer, R., Strobl, S.
Format: Conference Proceeding
Language:English
Subjects:
Agent
Auction
Conference management
Deciusion Support Algorithm
Engineering management
Information management
Information technology
Knowledge management
Negotiation
Ontologies
Ontology
Project management
Protocols
Technology management
Virtual enterprises
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper describes an application of a distributed knowledge-based multi-agent architecture to support automated negotiation between and within virtual enterprises (VE). Considering constraints of traditional negotiation approaches (predetermined protocols as well as hard-coded and non-adaptive strategies), we combine the agent paradigm with ontology-based solutions to overcome these limitations. Our negotiation ontology, used by agents, acts as a general framework that defines the basic terminology, interaction and protocols enabling agent to understand each other and to reach an agreement. Our multi-agent architecture consists of two types of agents that have rule-based behaviors. One is the Auctioneer that starts, monitors and ends an auction process in the name of a user that offers a particular product/service. Every new user who joins an auction has a configured MarketAgent that takes a participant role bidding for a product/service. To help an agent in its decision, a special decision support algorithm is implemented to recommend the best negotiation strategy, based on the user interests. We also describe the initial implementation of the system using JADE and JESS.
DOI:10.1109/FITME.2009.141