Based Hybrid Algorithm For Job Shop Scheduling Problem

Gradually as the global economic integration, the development, design and manufacturing cycle of the products are greatly shortened. One-piece, multi-variety and small batch production methods are becoming the mainstream of manufacturing production. This mode of production requires manufacturing resources can be scheduled and made use of in high horizon, and also requires the factors of production can respond quickly to market demand. Related statistics show that during the production process, 95% of the time of the parts is in transporting or waiting and other non-value added process. How to effectively plan the development of production, and rationally allocate productive resources, thereby reduce non-value-added part of the time, shorten the production cycle, reduce inventory cost, has been concerned by an increasing number of enterprises and research institutions .Nowadays, in the era that market is rapidly changing, it is crucial important to reduce the production cycle in order to respond quickly to market demand, for this is related to the enterprise's survival. Therefore, How to schedule the production resources and reduce that non-value-added processing time is the severe problem which every productive enterprise is facing.In response to this problem, this thesis explores into the job-shop scheduling problem based on the hybrid genetic algorithm. The hybrid algorithm, constituted of the mixed-use of the genetic algorithm and ant algorithm, aims to solve the problem of optimal solution. The hybrid algorithm, respectively, combining the advantages of both intelligent algorithms, effectively avoids the disadvantages of using only one algorithm to solve problems. Algorithm is characterized by processes based on the encoding method and the introduction of neighborhood search based on the way of variation. At the combination point of the two algorithms, the genetic algorithm is artificially increased in order to have a better solution path of the value of initial pheromone.This thesis takes minimizing the flow time of manufacture and minimum tardiness time as the optimization index, and uses the hybrid algorithm for solving the flexible job shop scheduling problem. This thesis designs the decoding method which can produce dynamic scheduling, selects several textile machinery parts which constitute examples of job shop scheduling problems, and uses different optimization index to solve this problem. The simulation results verify the hybrid algorithm for solving the flexible job shop scheduling problem with a faster speed and better global search capacity when compared with the genetic algorithm alone or ant algorithms.Finally, a prototype of flexible job-shop scheduling system is designed base on the hybrid genetic algorithm. This sysem initially realizes the functional indicators of shop scheduling by taking minimizing the flow time and the minimum delay time as the optimization index, and the scheduling results are dispatched by the form of the tables and the Gantt charts.