Patterns, 4D Simulations, and Artificial Intelligence–Driven Insights: Redefining Construction Workspace Management

Abstract

Workspaces in construction are more than just physical areas; they are critical resources that are shared among different crews. Inadequate planning of these spaces can have undesirable consequences, such as overlapping work areas among the crews, which can lead to conflicts that negatively impact productivity. Traditional models often fall short in providing a proper understanding of workspace needs and in establishing a variety of spatial-temporal plans to conduct the work. Recognizing the need for a more adaptive approach, this study uses a design science research methodology to present a four-dimensional (4D) simulation model that tests and analyzes different scenarios of performing activities based on patterns of movement. This study also presents a deep learning (DL) framework, combined with a space management dashboard to enhance decision making by predicting spatial conflicts and optimizing resource allocation. The simulation model demonstrates potential gains of up to 61.5% in reducing spatial conflicts. Additionally, the DL model achieved an accuracy of 98% in predicting potential conflicts, which emphasizes the role of data-driven approaches in construction management. This innovative approach highlights the role of advanced simulation and predictive modeling in understanding and optimizing workspace management, ultimately fostering more efficient construction environments.