Distributed Coordination of Resources via Wasp-like Agents

Abstract

Agent-based approaches to scheduling have gained increasing attention in recent years. One inherent advantage of agent-based approaches is their tendency for robust behavior; since activity is coordinated via local interaction protocols and decision policies, the system is insensitive to unpredictability in the executing environment. At the same time, such "self-scheduling" systems presume that a coherent global behavior will emerge from the local interactions of individual agents, and realizing this behavior remains a difficult problem. We draw on the adaptive behavior of the natural multi-agent system of the wasp colony as inspiration for decentralized mechanisms for coordinating factory operations. We compare the resulting systems to the state-of-the-art for the problems examined.