Design of an Agent-Based, Multi-user Scheduling Implementation
Selected Papers from the 4th Australian Workshop on Distributed Artificial Intelligence, Multi-Agent Systems: Theories, Languages, and Applications
Is it an Agent, or Just a Program?: A Taxonomy for Autonomous Agents
ECAI '96 Proceedings of the Workshop on Intelligent Agents III, Agent Theories, Architectures, and Languages
Hi-index | 0.00 |
In this paper, we present the design and implementation of a distributed multi agent meeting scheduling system. The system includes two types of agents: personal agent, one for each user, to negotiate on behalf of its owner, and a location agent to arrange suitable meeting rooms for the meetings. Personal agents have their own calendar and preferences. Invitees are classified as either important, without whom a meeting cannot be held or regular, whose absence does not prevent a meeting from being held. An agent has the ability to produce a counter proposal when it cannot accept a proposed date due to a conflict with its calendar, thus reducing negotiation time. Both types of agents can carry out several negotiations concurrently. We've made experiments to show how scheduling time and efficiency of the negotiation protocol vary with different calendar densities and different numbers of meetings.