Quantitative analysis of warnings in building information modeling (BIM)
Academic Article
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
2014 Elsevier B.V. All rights reserved. Building information modeling (BIM) provides automatic detection of design-related errors by issuing warning messages for potential problems related to model elements. However, if not properly managed, the otherwise useful warning feature of BIM can significantly reduce the speed of model processing and increase the size of models. As the first study of its kind, this study proposes to apply the Pareto analysis to investigate BIM warnings in terms of type and frequency. Based on warning data collected from three California healthcare projects, the analysis revealed that the 15-80 rule applies across the case projects and their design phases - 15% of the warning messages are responsible for nearly 80% of the warnings. Two other noteworthy findings include the following: (1) only the schematic design phase indicates a different Pareto rule of 25-80, as well as warning pattern from other design phases due to its unique purpose; and (2) the decisions of individual design teams are a major variable in the pattern of warning types. Lastly, time estimation for warning corrections is proposed based on learning curve theory to support efficient BIM warning management practices. The results and warning classifications presented in this study are expected to contribute to the design management and modeling practices of design teams involved in large, complex projects.
