Feasibility of Using New Technologies and Artificial Intelligence in Preventive Measures in Building Works

Abstract

The construction sector represents approximately 13% of global gross domestic product (GDP) and over 5% in Spain, employing more than one million workers. Despite its economic importance, the sector exhibits low digitalization levels and persistently high accident rates, contrasting with other industries that have successfully integrated digital technologies for safety improvement. Objective: This study evaluates the technical, operational, and regulatory feasibility of implementing digital tools and artificial intelligence (AI) in occupational risk prevention (ORP) within the Spanish construction sector. It focuses on identifying applicable technologies, assessing professionals’ perceptions of their practical utility, and analyzing key implementation barriers. Methodology: A mixed-method approach was employed in four stages: (1) a systematic literature review of digital safety tools; (2) a survey of 97 construction professionals using purposive sampling and validated through pretesting (Cronbach’s α = 0.82); (3) an analysis of official accident statistics; and (4) expert consensus using the Delphi method (three rounds, 75% consensus threshold). Results: Virtual reality (VR), augmented reality (AR), and mixed reality (MR) applications were identified as highly beneficial for training and awareness, with 78.2% of professionals supporting their use for safety training. Building Information Modeling (BIM) and drones were highlighted as the most valued tools for risk management and site supervision. Main implementation barriers include a lack of digital skills (35%), insufficient budget (30%), and high tool costs (25%). Contribution: This study proposes a mixed-method methodological framework—quantitative and qualitative—adapted to national contexts and validated through a Delphi consensus process. The framework prioritizes key technologies and identifies targeted strategies to overcome critical implementation barriers. © 2025 by the authors.