Automated quality control of building surfaces from 3D point clouds

Purpose This paper presents a fully automated methodology for geometric quality control of building surfaces using 3D point clouds. It addresses planarity, inclination and deviation analyses and generates reports compliant with current standards. Design/methodology/approach Raw point clouds are segmented into floors, walls and ceilings by analyzing orientation and elevation. This segmentation enables automated assessment of surface flatness, wall verticality and geometric deviations without manual intervention or BIM reliance. The method produces detailed visual and numerical quality control reports and is validated on synthetic and real datasets. Findings The approach achieves minimal algorithmic error compared to scanner precision and significantly reduces inspection times, analyzing about 100 m2 in under two minutes, outperforming other 3D scanning-based techniques. It identifies surfaces exceeding tolerance limits and accurately quantifies analyzed areas. Research limitations/implications Currently optimized for planar geometries per DIN 18202, future work aims to extend the method to complex architectural forms and optional BIM/HBIM integration. Practical implications Automating the quality control process reduces manual labor and inspection time, yielding cost savings and enhanced efficiency in construction workflows. Social implications Improved surface quality assurance promotes safer, more durable buildings, reduces maintenance costs and enhances occupant well-being. Originality/value The proposal introduces a fully automated, end-to-end workflow for quality control of all building surfaces without relying on BIM, advancing current quality control practices. The modular approach facilitates adaptation for different standards and complex geometries.