A text mining approach to performance enhancement in BIM pervasive major project delivery

Purpose: With regard to the Royal Society’s report Dynamics of Data Science Skills published in 2019, the UK Government policy paper National Data Strategy issued in 2020, and current research into and practice on building information modelling (BIM) underpinned digital transformation in the construction sector as well as big lessons learned from practices on major construction project delivery, this paper aims to present a new evidence-based learning experiment using text mining technique to inform both research and practice in the process to adapt to uncertainty, error, and the unexpected in BIM pervasive major project delivery.
Contents: This paper consists of three interconnected parts, including a systematic literature review to clarify the need for text analysis to tackle challenges identified from lessons learned on key issues including budget, quality and schedule in BIM pervasive major project delivery; a generic text mining process framework to enable the optimization of data collection and usage towards informed performance enhancement; and a structured discussion using findings from a text mining experiment on key technical elements of this framework to demonstrate its efficacy for problem identification and solving for BIM pervasive major project delivery.
Implications: It is expected that this paper can timely inform both academic research and professional practice in relation to applied text analysis as a technical solution to tackle identified challenges due to uncertainty, error, and the unexpected in BIM pervasive major project delivery. For academic research into text analysis in the subject field of the built environment, this paper initiates a new process framework with experimental studies for theoretical development. For professional practice in BIM pervasive major project delivery, this paper provides a novel technical solution for efficacious performance enhancement supported by evidence-based learning in a new scale of technical depth and breadth.