Abstract: Distributed flexible job shop scheduling is an important branch of production scheduling. As a common disturbance in the actual production, the dynamic arrival of jobs further increases the complexity and uncertainty of the job shop scheduling problem. Aiming at the distributed flexible job-shop scheduling problem with dynamic arrival of jobs (DA-DFJSP), a batching scheduling strategy was proposed, which transformed the original dynamic scheduling problem into a series of static scheduling problems over continuous scheduling intervals, and a mixed integer programming model was constructed with the maximum completion time as the optimization objective. On this basis, combined with the characteristics of the problem, the genetic algorithm was improved by the four-layer chromosome coding of batch, factory, process and machine, as well as a decoding method of fast greedy search insertion. At the same time, a variety of crossover and mutation operators were introduced to enhance the diversity of chromosomes. Finally, based on the FJSP standard example, a DA-DFJSP test case was constructed for simulation comparison experiments to verify the solution advantages of the proposed strategy and improved algorithm. The results show that compared to the traditional rescheduling strategy and the pre-improved genetic algorithm, the scheduling scheme proposed by the batched scheduling strategy and the improved genetic algorithm (IGA) has shorter completion period, more uniform plant processing load, and higher equipment work efficiency. There is a high degree of compatibility between IGA and the batch scheduling strategy, which can effectively improve production efficiency.