Detecting intrusion in hazardous areas is one of the priorities and duties of safety enhancement. With the emergence of vision intelligence technology, hazardous-area-detection algorithms can support safety managers in predicting potential hazards and making decisions. However, because of the dynamic and complex nature of the jobsite, high-risk zones have a different geometry and can be changed following the schedule and workspace of activity. This leads to hazardous areas being annotated manually. Thus, this study proposes a computer vision and a 4D BIM-based approach for intrusion detection in hazardous areas, called IDC4D. IDC4D comprises three modules: (1) the 4D BIM-based safety planning (4BSP) module, which analyzes the hazardous area; (2) the hazardous area registration (HAR) module, which delivers the hazardous area from the BIM model to the camera’s first frame image; and (3) the hazardous-area-intrusion-detection module (HAID), which applies the computer vision algorithm to identify the correlation between workers and hazardous areas. The efficiency of the IDC4D approach is validated by testing a maintenance project on the construction site. IDC4D supports the planner in choosing the plan and detecting the event of workers entering hazardous areas while working. It showed an average precision of 93% and 94% in phase 1 and phase 2, respectively. These findings provide insight into how varying geometries of diverse hazard areas can be handled for enhancing intrusion detection.

Performing risk assessments in construction requires collecting and analyzing project data and historical safety accident data, which is challenging due to the inherent complexities and dynamic nature of construction projects. To address these challenges, building information modeling (BIM) has been leveraged as a centralized digital repository that integrates data and provides a holistic 3D view of a project. Previous studies have highlighted BIM’s significant functions for risk assessment, such as visualization, simulation, and clash detection. However, these studies often overlook the incorporation of temporal information, which is crucial for assessing risks accounting for the dynamic conditions of construction sites. This study develops a 4D BIM-based risk-assessment framework by integrating spatial and temporal data to respond to dynamic site changes. The framework leverages 4D BIM to combine 3D model data with time-, resource-, and logistics-related information, enhancing the tracking and evaluation of construction progress. The study involves investigating major construction accidents, classifying their risk factors, establishing risk-factor identification algorithms, and implementing the framework on a web-based platform for validation. This approach offers a comprehensive risk-identification strategy, applicable to multiple accident types, with intuitive visualization using BIM models, benefiting from managers’ experiential knowledge and enabling effective risk assessments and mitigation strategies. Consequently, potential safety risks at construction sites can be efficiently identified using interconnected spatial and temporal data while tracking changes in risk levels in real time and visualizing them on a web-based platform.

The construction industry has long been recognized for its complex safety regulations, which are essential to ensure the well-being of on-site employees. However, navigating these regulations and ensuring compliance can be challenging due to the volume and complexity of the documents involved. This study proposes a novel approach to extracting information from construction safety documents utilizing Large Language Models (LLM), called CSQA, to provide real-time, precise answers to queries related to safety regulations. The approach comprises three modules: (1) the construction safety investigation module (CSI) collects safety regulations for building the information needed. By leveraging a collection of safety regulation PDFs, the system follows a process of text extraction, preprocessing, and global indexing for efficient search. (2) The safety condition identification module (SCI) retrieves the CSI database; after that, the LLM, with its extensive training, processes user queries, searches the indexed regulations, and retrieves pertinent information. (3) the safety information delivery (SID) would provide the answer to the user and incorporate a feedback mechanism to further refine system accuracy based on user responses. Preliminary evaluations reveal the system's superior performance over traditional search engines, owing to its ability to grasp query context and nuances. The CSQA presents a promising method for accessing safety regulations, with potential benefits including reduced non-compliance incidents, enhanced worker safety, and streamlined regulatory consultations in construction