Research On Assembly Job Shop Scheduling Problem With Uncertainty Processing Time Based On Genetic Algorithm

With the development of the times,multi-specification and small-batch production has become so mainstream.Customer demands are increasingly diverse,and competition between business is also becoming more fiercer.For this reason,Enterprises must constantly promote refinement and normalization of job shop operations management so as to be sure of success in the fiercer competition.At the same time,managers must pay attention to the random disruption events such as uncertain processing time in production and processing,and consider the influence of uncertainty factors on production before the scheduling strategy is formulated,thus making the corresponding scheduling plan and adjusting it constantly in operation to adapt to these possible uncertainties.There is still a large part of production and assembly process that cannot be fully automated and mechanized in the contemporary production job shop,especially in the assembly job shop,many processes to be completed by manual operation of workers.Due to the discrepant processing proficiency of different workers,the completion time of workers' operation procedures,influenced by many factors such as their physical condition,mood and work environment changes,is also not the same,which inevitably lead to the uncertain processing time.The uncertainty of processing time is a kind of main uncertainty factor which can affect the execution effect of the whole scheduling scheme.As a result,it may bring about the earliness or tardiness completion of products.In the current just-in-time production,whether earliness or tardiness will bring extra costs to the enterprise and increase the risk of default,which is try to be avoided.The process constraint of assembly job shop scheduling problem is more complicated and it may exist in multiple assembling stages,so it is much different from the traditional production job shop scheduling.In this thesis,the assembling constraint and assembly stage are taken into consideration with the assembly job shop scheduling.In order to solve this problem with uncertain processing time existing in the real situation,this thesis by introducing the fuzzy numbers theory,uses the triangular fuzzy number to represent the processing time of the procedures and sets up the mathematical model of assembly job shop scheduling in uncertainty,and also through designing adaptive genetic algorithm based on feasible region search to solve it.Then,colsidering that lots of assembly job shops are flexible ones,this thesis establishes the scheduling model in uncertain processing time for more complex flexible assembly job shop,and solves this problem by improved genetic algorithm.Finally,computations are designed respectively for the general and flexible assembly job shop scheduling problems under the uncertainty of processing time,the solutions finally carried out by the simple genetic algorithm and the improved genetic algorithm,which show that the algorithm proposed in this thesis can efficiently solve this scheduling problem through analyzing.