Abstract: In view of the serious waste of package materials in domestic logistics compared with foreign logistics, and the situation of the huge requirement of packing optimization, a new type of bin packing scheme is proposed, and a packing optimization method based on Genetic Algorithm is designed. The objective is to select a package specification with minimum surface area but can accommodate a number of cuboid-shaped items from existing package specifications. In the method, genetic codes involving packing sequence and orientation, along with a space segmentation method, is used to find a packing solution. Considering the high cost of traditional laser scanning measuring equipment, a box dimension measurement method based on RGBD is proposed to adapt to the management mode of domestic storage, and to provide dimension of items for packing optimization. Experiment results demonstrate that the proposed method is effective on finding package specification with minimum surface area and can measure box dimensions quickly and accurately.

Key words: genetic algorithm, RGBD, box measurement, packing optimization, calibration