Abstract: Aiming at the scheduling problem of workpiece process with complex partially ordered relationship in garment job shop, the characteristics and difficulties of this kind of problem are analyzed, and the improved arithmetic optimization algorithm is used to solve the complex partially flexible job shop scheduling model with the optimization goal of minimizing the maximum completion time. In the algorithm, the multi constraints of decoding are fully considered, and the optimal processing time of subsequent processes is determined through the previous process of the workpiece and the available time period of the machine. In the exploration stage, a two-dimensional clustering based on fitness and process completion difference is designed, and the population diversity is expanded through inter class crossover and effective mutation strategies, In the development stage, a hierarchical neighborhood strategy is designed for bottleneck work orders and bottleneck machines to enhance the local search ability of the algorithm. Through simulation experiments, extended benchmark examples are used to verify the superiority of the improved arithmetic optimization algorithm based on hierarchical neighborhood strategy in partially ordered flexible job shop scheduling.