Visual log analysis method for designing individual and organizational BIM skill

Tsukasa Ishizawa, Yasushi Ikeda

Keywords: Building Information Modeling, log mining, visual analytics, skill design, intercomparison, self-monitoring

TL;DR: Visual analytics and clustering extended the utility of BIM log mining, which provides four cluster types representing activities’ commonality and the level of contribution.

Abstract: Extending the existing Building Information Modeling (BIM) log mining approach, the paper proposes a novel method for visual analytics on clustered logs collected from multiple organizations. Implementing multiple datasets as case study processes demonstrated that the analytics effectively visualized the structured BIM events in command, organization, and user layers. Identified four major cluster groups correspond to logs' commonality and level of contribution significantly helped decipher recorded BIM activities without requiring background knowledge. The proposed technique allows the intercomparison of BIM activity that enables data-driven skill design for individuals and organizations. Such monitoring allows BIM users and teams to respond to transient project situations dynamically, which is expected to mitigate the observed major BIM problems of skill and management. The method contributes to increasing the likelihood that the model will be used throughout the project duration, leading to enhanced BIM's expected value in projects.