Including Measurement Uncertainty in a BIM

Building Information Modeling (BIM) can be defined as the generation and management of digital models of large structures, such as buildings and other infrastructures, throughout their sometimes decades-long lifecycle. Lidar has become a popular tool to document changes in the geometry of a structure during its lifecycle.

For large structures many scans (lidar depth maps) must be combined to achieve complete coverage; however, combining the scans required to model the geometry of large, and sometimes complex, features may introduce geometric distortion as a result of cumulative alignment error. Moreover, changes in scanning technologies and methodologies can increase the deviation between the geometry of digital representations of the original structure and digital models generated later in the structure’s lifecycle. The use of best practices makes it possible to create digital models generated in accordance with accepted standards using measurements with quantified uncertainty values. Best practices make it possible to distinguish between variation due to structural changes and variation due to measurement uncertainty due to the scanning technology and methodologies employed.

For our test study, we generated a digital model of a 130 m x 55 m x 20 m stone-and-brick building over a steel structure using a combination of lidar systems, laser trackers, spherical artifacts, and contrast targets. We built a reference model with an expanded uncertainty (k=3) of 33.63 mm, a one-sigma surface RMS of approximately 2.36 mm, and an average spatial resolution of about 20 mm. By including measurement uncertainty in our digital model we will be able to monitor the structure for decades to detect changes in physical geometry, such as changes due to local seismic activity or structural degradation.


David MacKinnon, Ph.D., P.Eng.

Research Officer / Agent de recherche

Mechanical Metrology / Métrologie mécanique

Measurement Science and Standards / Science des mesures et étalons

National Research Council of Canada / Conseil national de recherches Canada

One comment

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.