Methodology And Application Of Production Scheduling Based On Multi-Agent System

Multi-agent system is one of the important research fields of artificial intelligence. It has particular predominance for solving complex distributed problems, which is an important method full of applying value. Production scheduling is at the middle layer in the CMIS system structure, which is the joint of control and management. On the one hand, it will supply decisions for the enterprise; on the other hand, it will arrange production tasks and supervise the control layer. So the production scheduling is the key of the CIMS. Because the most scheduling problem is NP hard, it is impossible to find out common algorithm with polynomial complexity. We combine the multi-agent technique and production scheduling and decompose complex task to units by the distributed characteristic of multi-agent system. We reduce the complexity of system designing through agents' cooperation.In this paper, we have analyzed the concept and action of multi-agent system, supposed a scheduling model based on multi-agent system. We also simulate a flexible job-shop scheduling problem through software, which is object oriented. The main research work is described as follow:1) We discuss the basic concept, characteristic and classification of an agent, analyze the communication and cooperation among agents.2) We set up workshop MAS scheduling model by the discussion to agent oriented analysis and modeling. In this model, we adapt entity agent and abstract the entities in the workshop to manager agent, task agent and resource agent. The agents cooperate each other through contract net and complete the production order. We build and analyze the three agents' functionalities, discuss the cooperation process and make a simulation of a flexible job-shop scheduling problem.3) We suppose a dynamic scheduling model based on workshop MAS scheduling model and make simulations of machine uncertainty and task uncertainty. Additionally, we have introduced a method to solving job-shop scheduling problem based on modified genetic algorithm and supposed the mixed scheduling model combining genetic algorithm.4) We apply genetic algorithms, neural network algorithms and multi-agent technique to production scheduling by summarizing the past research achievements.We also introduce the structure and functionality of production planning and scheduling software package acknowledged by Zhejiang province's science and technique project. By analyzing the system functions, we suppose the whole structure of the software package and discuss the system structure from two different views of function and workflow. At the same time, we explain the database and software designing from software engineering. At last, we make a scheduling to an actual production problem by using the software package and analyze the scheduling results.