This study aims to identify key factors influencing digital twin implementation in construction and to validate their interrelationships using partial least squares structural equation modelling (PLS-SEM).
A structured, multi-phase research approach was adopted. Key constructs were identified through literature review and expert input, followed by the design of a questionnaire administered to 436 construction professionals. Exploratory factor analysis was used to refine constructs, and PLS-SEM was applied to test relationships. Second-order constructs were modelled to evaluate the integration of computer vision (CV), Internet of Things (IoT), digital technology integration and data-driven project performance within a unified framework.
The results demonstrate that CV–IoT integration, digital technology adoption and data-driven performance significantly contribute to real-time construction monitoring. Key relationships – such as CV–IoT to edge processing (ß = 0.754), digital integration to labour, material, equipment and activity (LMPA) monitoring (ß = 0.832) and data-driven performance to real-time monitoring (ß = 0.748) – confirm the model’s strength. These findings underscore the value of integrated digital systems in enhancing site visibility, progress tracking and predictive decision-making.
The model identifies critical digital and operational factors essential for structured digital twin prototype development, enabling real-time monitoring of LMPA to support automated tracking and improve cost, time and resource efficiency in construction projects.
This study develops a PLS-SEM-based digital twin framework that systematically integrates CV–IoT, edge processing, building information modelling and data-driven performance. The model provides a structured understanding of how these technologies collectively enhance productivity monitoring, progress assessment and project control in construction.
