Research On Flexible Job Shop Scheduling Based On Improved Genetic Algorithm

With the development of China from a strong country in manufacturing to a manufacturing power,energy saving and efficiency increasing has become the only way to upgrade and transform the modern manufacturing industry.At this stage,it has a strong practical significance to reduce energy consumption in manufacturing industry.Considering the background of modern multi variety,less batch manufacturing mode,energy-saving and consumption reducing,the flexible job shop of modern production mode is taken as the research object of this thesis,reducing the completion time and energy consumption is taken as the research object of this thesis,the improved hybrid genetic algorithm and the improved non dominated genetic algorithm are taken as the methods for the single objective and multi-objective scheduling problems.The reliability of the algorithm proposed in this thesis is verified by various classical examples and actual production data.First of all,the types,characteristics,representation methods and general objectives of flexible job shop are systematically studied and elaborated in this thesis,and the solutions of single objective and multi-objective flexible job shop scheduling problems are also summarized.Secondly,in order to solve the problem of poor local search ability and low search efficiency in the late evolution of genetic algorithm,an improved hybrid genetic algorithm is proposed in this thesis.This algorithm makes up for the disadvantages of the single genetic algorithm which has poor local search ability and the tabu algorithm which relies too much on the initial solution.The improved global search strategy is used to generate the initial population,and the quality of the initial population can be effectively improved.By improving the priority process crossover method,the situation that the individual gene value of the crossed offspring is the same as that of the parent is effectively avoided.In the tabu algorithm,the key process is disturbed to produce disturbance,which effectively reduces the useless neighboring solutions and improves the calculation efficiency.The availability of the algorithm is verified by the MK example,and compared with different algorithms in other literatures,the algorithm in this thesis has achieved better solution in solving the MK example.Compared with the iterative curve of the standard genetic algorithm,it shows that the algorithm,in this thesis,has a significant improvement in the efficiency of optimization,population quality and stability of reconciliation.Through the application of two examples,the results of this algorithm are better than those of other literatures,and the advanced of the algorithm is proved.Finally,on the basis of the single objective flexible job shop scheduling model in Chapter 3,by analyzing the production energy consumption composition of modern flexible job shop,the calculation methods of various energy consumption are determined,and the machine start-up and stop constraints and no-load constraints are added,the scheduling model aiming at the shortest completion time and the least workshop energy consumption is established.The non dominated genetic algorithm is designed with the idea of improving the population initialization and genetic operation in the third chapter.The validity of the scheduling model and algorithm established in this thesis is verified by the actual production case data.