Modeling And Simulation Of Flexible Job-shop Scheduling System Based On Contract Net Protocol

Job-shop scheduling problem, because of its NP-Hard features, with the growing-scale of the problem, traditional centralized solutions is growing bloated when solving these rising size of problem. And in a more realistic environment of the flexible manufacturing, it is unable to meet the complicated, dynamic and stochastic demand. In this context, many scholars begin to study the introduction of a more suitable model of the adaptive system based on autonomy and coordination. This paper analyzes the modern domestic-related and foreign-related research on the flexible job-shop scheduling problem, and takes the flexible production and processing of discrete operating manufacturing plant as a research object, and then establishes an agent-unit autonomic and consulted model based on contract net protocol. In this model, it introduces an algorithm called Q-Learn and a valuation-process algorithm for the local optimization. It is an effective and practical solution for the flexible job-shop scheduling problem.The main content of this paper can be summarized as follows:1. First of all, briefly summary and analyze the researches in recent years, and then point out the significance of this paper.2. To the complexity, uncertainty, multiple objectives, co-ordination and multi-constrained of the problem, this paper proposes a real-time scheduling framework model based on the agent unit and contract net protocol under the guidance of the complex adaptive system(CAS). And then analyzes the advantages and disadvantages of the system, points out the contract net protocol in the system lacks of optimization and dynamic learning ability.3. Embeds the Q-Learn algorithm to the tender evaluation and decision making system, and then introduces the detailed definition and description of the model based on Q-Learn algorithm, gives the contract net protocol dynamic intelligent learning ability.4. Proposed the concept of process value in the tender evaluation process, quantifies the system goal to the form of "value", then attaches it to the sub-tasks to optimists the local optimization of the system.5. Finally uses the tools of Java, Swarm and Matlab to design and develop a comprehensive simulation system. And takes a simulation test for the viewpoint which proposed in the paper. It makes a feasibility try for the theoring the practical application of the problem.This paper proposed a program to solve the job-shop scheduling problem in the flexible manufacturing environment. To some extent, it moves the solving method of job-shop scheduling problem forward a step.