Establishing Point Cloud Data Requirements for Target Level of Development (LOD) Standardization using Plannerly

Abstract

Building Information Modelling (BIM) and Scan-to-BIM workflows for existing buildings routinely specify target Level of Development (LOD) classes such as LOD 300 and LOD 350, yet they rarely define the measurable point cloud parameters, including registration root-mean-square error (RMSE), point spacing, density, and deviation tolerance, required to achieve them. The objective of this study was to define quantitative requirements for point cloud data for the standardization of the LOD target value and to integrate these into a BIM Execution Plan (BEP) based on Plannerly. An interior corridor was selected as an experimental case study and scanned using the Leica RTC360 and BLK360 terrestrial laser scanners. Four quality levels (T1 to T4) were generated by varying the registration accuracy and point spacing, and the reference geometry was modeled in Autodesk Revit. Each level was compared to the model in CloudCompare using the Cloud-to-Mesh analysis, and conformity was evaluated using 95th percentile deviation metrics (D95) aligned with the tolerance bands of the Level of Accuracy (LOA). The results showed that LOD 300 was achieved by T1, T2, and T4, while T3 failed; LOD 350 was achieved by T1, was borderline for T4, and was not met by T2 or T3. These results were summarized in a standardized requirements matrix that links measurable point cloud parameters to the target LOD classes . The matrix was embedded in Plannerly as a structured BEP requirements set and provides a reproducible quality control workflow for the specification and verification of scan-to-BIM projects.