An adaptive scheduling system in knowledgeable manufacturing based on multi-agent

Abstract

To address the uncertainty of production environment in knowledgeable manufacturing system, an interoperable knowledgeable dynamic scheduling system based on multi-agent is designed, in which a knowledge representation with a series of problem characteristics for various scheduling problems is adopted and problem-based function modules are constructed by using agent technology. An adaptive scheduling mechanism based on Q-learning (known as MA-Q policy) is proposed for guiding the equipment agent to select scheduling strategy in a dynamic environment. Simulation experiments show the scheduling system is of high intelligence and interoperability, and can constantly adapt to environmental changes by self-learning.