Research On Multi-variety And Small-batch Hybrid Flow Shop Dynamic Scheduling Based On IGA

With the continuous development of "Made in China 2025" and "Internet+",diversified and rapidly changing customer demands have become the key driving force for the transformation and upgrading of manufacturing.The production mode is gradually changing from large-scale production for inventory to multi-variety and small-batch production for orders.The hybrid flow shop can make full use of production resources,reduce costs,and improve profitability.However,there are common problems in processing workshops,such as unreasonable scheduling of singlepiece production and batch production,untimely handling of dynamic disturbance events,resulting in prolonged processing times,and poor production management.Therefore,taking the hybrid flow shop under the small-batch and multi-variety production mode as the research object,this paper studies single-piece production scheduling,batch production scheduling,and batch production dynamic scheduling problems,respectively,and designs corresponding scheduling methods based on genetic algorithms.First,for the scheduling problem under single-piece production,a mathematical model for hybrid flow shop scheduling with the optimization goal of minimizing the maximum completion time is established,and an improved genetic algorithm is designed to solve it.The algorithm used a vector coding method based on the job processing sequence in the first process and decoded it according to the principles of first-come first-processing,first-idle first-processing,and job readjustment.A hybrid initialization strategy is designed to improve the quality of the initial population.Multineighborhood search is introduced to perform local search on the current optimal solution.The effectiveness of the improved genetic algorithm(IGA)is verified through processing workshop examples,providing a theoretical reference for single-piece production scheduling.Secondly,for the scheduling problem under batch production,a mathematical model for hybrid flow shop scheduling with lot streaming based on equal batch splitting is established.The batch processing is carried out using equal batch splitting,nonmixable sub-batch processing,and no-wait sub-batch transfer.The objective function and constraints of the model are established based on sub-batch constraints.Combined with the characteristics of sub-batches,the genetic algorithm is further improved for solving.During decoding,the sub-batch priority principle is followed,and a mixed initialization strategy including sub-batches is designed.Finally,the effectiveness of the method is verified through a customized connecting rod batch processing production example,providing some reference for batch production scheduling.Finally,for the dynamic scheduling problem with lot streaming,a mathematical model for hybrid flow shop dynamic scheduling based on rolling windows is established.Firstly,based on the rolling window technology,rolling windows and job sets are constructed to transform dynamic scheduling into static scheduling for initial scheduling and rescheduling.A rescheduling strategy and is designed for dynamic disturbance events such as urgent jobs,machine failures,and newly arrived jobs.The improved genetic algorithm is used to reschedule the jobs in the scheduling window.Through the connecting rod batch processing process in a dynamic environment,the effectiveness of this method in solving various disturbance events is verified.