Design And Implementation Of Flow Shop Production Schedule System

As the development of economy, market competition is becoming fiercer and fiercer.,To the manufacturing enterprises that want to win this competition, it is one of the key factors whether they can optimize the resource allocation to cut the product period and maximize the product. Most product activities are performed in workshops. As a result, it will help improve product performance through distributing resources in reason and cutting product period to maximize the product in workshop level. However product scheduling is the very important part of workshop management, so product scheduling in workshop become one of hot topics in manufacturing research and is meaningful in both academic and reality.With the effort to the workshop product scheduling management system, the author provides a product scheduling management system that is fit for flow shop environment which is also a solution.Through surveying business processes of diode workshop in Company Goodark, analyzing bills and case data moving among processes, it is clear what is the flow shop scheduling problem and which characters the flow shop environment has. A genetic algorithm based on real number encoding is provided to solve this flow shop product scheduling problem. With this algorithm, an object oriented and multi-agents architecture software system is designed and implemented.Math formed description that confirms the input, output and evaluation standard of the workshop scheduling problem is provided. With the real number based genetic algorithm,a algorithm that meets the requirement of evaluation standard is provided through making sure the encoding rule, fitness function, choosing, crossing, variation operations and end condition. And it has been written to an independent software package.Based on the genetic algorithm above and the business requirement of diode workshop in Company Goodark, an object oriented and multi-agents architecture scheduling system is provided through high level design, detailed design and implement. Because the workshop system is divided to resource agent, scheduling agent and mission scheduling, the system could not only deal with daily order generated by MRP, but also cope with some urgency situations such as device failure by instant scheduling in order to maximize the product. This system has gone live which proves the theory in the paper.