Improved Routing Wasps for Distributed Factory Control

Abstract

Agent-based approaches to manufacturing scheduling and control are attractive because they offer increased robustness against the unpredictability of factory operations. Previously, we introduced a new approach to coordinating factory routing and scheduling based on a computational model of wasp behavior. The natural multi-agent system of the wasp colony is highly effective in self-organizing the allocation of tasks necessary to fulfill the needs of the nest and has proven useful as an effective model upon which to base our distributed approach to factory control. In this paper, we improve upon our original formulation of the routing wasp with the addition of a tournament of dominance contests among routing wasps competing for the same job. We experimentally evaluate this improved performance.