Theory and properties of a selfish protocol for multi-agent meeting scheduling using fuzzy constraints
Xudong Luo, Ho-fung Leung, Jimmy H. M. Lee · 2000
. This paper develops an agent-based methodology for meeting scheduling. In such a multi-agent system, each agent acts on behalf of a user. For each user the meeting scheduling problem is modeled by a fuzzy constraint satisfaction problem, and an appointment is made by negotiations among agents. A negotiation procedure concerns with two key components: the protocol for organizing negotiations among agents, and the operator for fusing agents' individual evaluations for a feasible time slot. In particular, we propose a kind of selfish protocol, and present an axiomatic framework for fusion operators. In addition, a meeting scheduling example is used to illustrate the proposed methodology. keywords: Scheduling, Constraint Satisfaction, Multi-Agent Systems, Distributed AI, Uncertainty in AI. 1 Introduction A meeting scheduling task usually requires a lot of efforts in communication and negotiation among attendants since they may have different timetables, constraints and preferences. If...