An acceptance sampling method, based on obtaining physical measurements, for accepting 2D building plans generated from 3D imaging data is evaluated and demonstrated using an actual GSA (General Services Administration) facility. The objective of the effort was to determine practical/feasible limits for the tolerances (T), sample size (n), and the maximum percent of the population (P) out-of-spec that is acceptable for GSA. A total of 285 measurements were obtained from a GSA 3D imaging project site by the National Institute of Standards and Technology (NIST) and Carnegie Mellon University (CMU). These measurements were then compared to corresponding measurements extracted from 2D plans or 3D models. Several thousands of simulations were run for different combinations of samples (i.e., to randomize the data sampling) for a given set of n, P, and T. The data collection, data analysis, and findings are presented in this paper.
Citation: Journal of Construction Engineering and Management-ASCE
Pub Type: Journals
3D imaging, acceptance sampling, 2D building plans, 3D model, quality assurance.