Digital Twin Construction: Feasibility of an Automated Closed Loop Planning and Control Cycle

Abstract

The scale and complexity of construction projects present production planners with significant challenges when making planning decisions concerning resource allocations, construction methods, and activity sequences. Among the challenges are a lack of project situational awareness and reliable tools for decision outcome prediction. Digital Twin Construction (DTC) aims to automate decision-making by incorporating information and monitoring technologies into a lean closed-loop planning and control system. We propose to implement and prove feasibility of a DTC-enabled closed-loop ‘Plan-Do-Check-Act’ planning and control cycle. To date, no-one has successfully demonstrated such functionality. A practical DTC system will be developed and tested in a controlled laboratory setting. The system will include: (1) Situation Awareness Recognition, enabling automatic data acquisition from the construction site and its interpretation; and (2) Planning Simulation, leveraging gathered situational insights and simulation technologies to identify optimal planning decisions. Based on the DTC system, various decision-making scenarios of increasing automation levels will be evaluated. Successful demonstration of such a workflow will establish a solid foundation for research and development endeavours, driving the industry towards more sustainable and efficient practices. The resulting technology will be applicable directly to enhance existing prefabricated and modular construction methods.

Research Aims and Objectives

In this research, we propose to explore the feasibility of the basic operation of the full cycle of automated ‘Plan-Do-Check-Act’ (PDCA) (Deming 2000), thus proving the feasibility of the DTC concept. While the benefits of data-driven closed-loop planning and control cycles have been formulated in theory, no such closed-loop system has yet been implemented or demonstrated on any scale. This is mainly due to technical, economic and commercial barriers. Moreover, the adoption of such systems requires a substantial shift in mindset and the development of new skills for construction professionals, adding to the complexity of implementation.

The primary aims of this study are to demonstrate the closed-loop data-driven production control workflow within the context of the emerging digital twin construction (DTC) environment, to establish the feasibility of decision-making augmentation, and to explore different levels of automation in such an environment. To achieve this, the following objectives will be pursued:

  • Design and build an experimental system to demonstrate a practical DTC solution for production planners in a small-scale experimental project in the laboratory.
  • Demonstrate the automatic process of data acquisition from the construction site, its interpretation into actionable information, and subsequent storage in a cloud repository.
  • Demonstrate the effectiveness and the efficiency of production planning that leverages predictive situational awareness derived from simulations.
  • Investigate various decision-making scenarios with increasing levels of automation.
  • Provide recommendations for application of closed-loop planning and control cycles in precast construction and conventional construction, and provide suitable prototypical hardware and software.