A PLS-SEM based framework for enabling digital twin systems for real-time monitoring and control in construction projects

Purpose 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). Design/methodology/approach 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. Findings 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. Practical implications 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. Originality/value 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.