The work aims to solve the key loading algorithm problem in the automatic loading and unloading of air luggage, realize automatic loading and unloading of air luggage and meet the actual needs of flow operation. Based on the key point loading strategy, an improved particle swarm optimization (PSO) algorithm that considered constraints such as the weight, volume and loading order of luggage was proposed with the utilization of loading space as the optimization objective. Firstly, the key point method was used to output all the point sequences of the bags to be loaded on the pipeline. Then, the speed and position of the particle swarm optimization algorithm were redefined according to the constraint conditions. The space utilization was used as the fitness function for iterative optimization, and the global optimal solution was output to optimize the loading position and attitude. In the experimental part, the real luggage data was used to verify the algorithm. The results showed that the improved particle swarm optimization algorithm could improve the box space utilization by 10.8% and the average layout efficiency by 26.5%. The proposed loading algorithm can effectively solve the actual luggage packing problem, and provide a theoretical basis and reference for the cargo loading of luggage flow process.