Turning a point cloud into a Building Information Model (BIM)

Defining and validating the accuracy requirements for existing buildings

Abstract

Digitization of existing buildings is one of the main future goals, leading to efficient planning, renovation and maintenance. Among the existing buildings, a significant share is protected as a cultural heritage and their management is supervised because interventions on the protected sites are limited. Building information modeling (BIM) provides the opportunity to integrate accurate as-built information into the digital environment where it can easily be accessed and used. A digital representation of building creation usually starts with the acquisition of spatial data (point cloud), which is then used to create a semantically enriched model with certain geometric accuracy (BIM). In order for the model to serve its purpose, it is important to define how accurate the model should be. Since there are currently insufficient definitions of geometric requirements for specific BIM use cases, the research hypothesis was that the quality of BIM greatly depends on the modeler. The identified issue was approached with a study case. Using the point cloud of the existing building, the BIM was made and validated based on pre-defined accuracy requirements. Different accuracy validation methods were used in the process. Based on the results of the study case, conclusions and recommendations for efficient BIM creation were prepared.