Research On The Application Of Genetic Algorithm In A Company's Parallel Multi-machine Workshop Scheduling Problem

In the context of intelligent manufacturing as an important strategic background of China's manufacturing 2025 plan,the fast-moving consumer industry has introduced the development pattern of overtaking in the corners,and the development of A company has reached the strategic stage of overall planning,coordinated layout,and model transformation.With the increase of production lines in the production workshop of A company,limited resources and other constraints,the existing production scheduling plan can no longer meet the production scheduling needs of the factory.Digital workshop scheduling solutions play a vital role in the smooth progress of factory production.A reasonable and effective production scheduling plan can greatly reduce the work of planners and avoid waste of capacity due to human error and unreasonable resource allocation.The scheduling problem of the ice cream packaging line of the A company has its unique characteristics.First,its production equipment is a non-identical parallel machine.secondly,each production line has different category priorities.The higher priority needs to be arranged in Produced together,and certain tasks have specified processing sequence requirements.Therefore,this research takes allocation and sequencing as the key decision,and takes the minimum total completion time as the scheduling goal.According to the actual needs of A company,a mathematical model of the parallel multi-machine shop scheduling problem of A company is established,and an improved genetic strategy is used to obtain a feasible scheduling plan.In the traditional genetic algorithm,the offspring produced by the crossover operation cannot well inherit the excellent characteristics of the parent.The crossover operation and the mutation operation may cause the hard-to-get good solutions to be destroyed or lost.In the case of more constraints,it is difficult to get a satisfactory solution quickly and effectively.However,the actual demand of the ice cream production scheduling workshop of A company is more constrained.Therefore,in order to better solve the scheduling problem of the ice cream product packaging line of A company,this research has made some improvements on the basis of traditional genetic algorithms.The excellent individuals in the initial population and the excellent individuals generated by the crossover are stored in the external solution set.The excellent individuals retained in the external solution set replace the individuals with the lowest adaptability after genetic manipulation,which can greatly avoid destroying the good solutions.and the crossover probability is set to a larger value of 0.99 to improve the operating efficiency of the algorithm and the species diversity,and generally increase the probability of getting a better solution.In order to verify the effectiveness of the algorithm in solving the multi-machine parallel workshop scheduling problem of A company,this study used 10 cycles of data for testing.The test results are compared with the solution results of the traditional genetic algorithm and the optimal solution obtained by CPLEX solver.Finally,it is concluded that the improved genetic strategy used in this paper is feasible and efficient in solving the multi-machine parallel shop scheduling problem of A company.