In the construction of large and complex frame-panel structures, it is difficult to measure the quality of the frame due to bending and torsion, which affects the panel installation. In addition, the traditional glass panel deepening design suffers from low efficiency and poor intelligence. Point cloud data (PCD) and heuristic algorithms provide a perfect solution for the status obtain of completed frame and intelligent of panel design. According to the Buildings of International Financial Forum located in Nansha District, Guangzhou city, the intelligent construction techniques for large-scale complex frame- panel structures based on PCD and heuristic algorithms were studied in this paper: Conversational modeling methods suitable for frames have been proposed in response to the design requirements of panels; The algorithms was proposed to extract the axis of the curved and twisted pip which have a polygon section to address the difficulty in skeleton geometry description; A unit-wise automatic extraction method has been proposed to improve the efficiency of panel partitioning; different strategies were used to generate panel layout schemes based on the Guillotine algorithm, and layout results were evaluated. The research show that the reverse modeling method can accurately obtain the frame model, the intelligent partitioning method for panel is simple and effective, and excellent cutting plans can be quickly obtained using heuristic algorithms. The results can be used as a reference for digital construction transformations of similar structures.