An optimized placement of building drawings with moving least squares and K-means clustering

In this paper, we propose a method that can efficiently process the arrangement of building drawings using K-means clustering and vector field constructed by optimized moving least squares (MLS). In the proposed framework, after selecting the area to actually place the buildings, the vector field is generated by optimizing the MLS based on this area, and the angle to rotate the building drawing is determined based on this field. In the simulation step, K-means clustering is used to determine the initial layout of the building drawings, and their locations are advected based on the vector field calculated by MLS to further locate new building drawings in the empty space. This allows a maximum number of building plans to be placed within a given area. The practicality of the proposed method was verified by comparing with the actual architectural design, and the efficiency of the overall design process was improved by greatly reducing the amount of time and work required.