The construction sector has long been known for its complicated safety requirements, which are critical to ensure the well-being of workers on site. However, interpreting these requirements and maintaining compliance can be difficult due to the amount and complexity of the paperwork involved. This leads to difficulty extracting safety information from requirement documents. Besides, information that is presented in a manner similar to human-like responses can improve employee understanding. This study proposed a Construction Safety Query Assistant (CSQA) approach to enhance the extraction and knowledge of construction safety regulations using Large Language Models (LLMs). CSQA comprises of three primary components: (1) the Construction Safety Investigation Module (CSI), which gathers and processes safety regulation documents through text extraction and preprocessing to build a searchable database; (2) the Safety Condition Identification Module (SCI), which utilizes LLMs to interpret user queries and extract relevant information from the CSI database, capitalizing on the models' ability to understand context and subtle textual nuances; and (3) the Safety Information Delivery Module (SID), which presents the retrieved information to users and integrates a feedback loop to refine the accuracy and relevance of responses based on user interaction. The CSQA approach was validated with 2 case studies that offered more contextually relevant, possibly lowering non-compliance risks, improving worker safety, and simplifying the consultation process in the construction sector. This study emphasizes its potential to transform access to crucial safety information in the construction industry.
The construction sector is known for exposing workers to numerous potential hazards, with falls from heights being the leading cause. These fatal fall accidents not only result in human loss but also impose significant financial costs on construction projects. However, current safety planning and management is typically carried out manually using safety documents and 2D models, which are time-consuming and labor-intensive. There is also a lack of visualization for the placement of temporary safety facilities (TSFs) during construction. Meanwhile, Building Information Modeling (BIM) has the potential to be used as a comprehensive workspace planning for TSFs in a scheduling manner. Therefore, this study proposes the development of a construction safety dashboard to inform workers about fall hazards using spatial–temporal data stored in 4D BIM. The proposed approach includes four modules: (1) identification and assessment of risk from identified work activities, (2) development of 4D BIM model, (3) creation of a dashboard to share safety knowledge, and (4) validation of the dashboard through interviews with safety managers and site workers. This approach is tested on a stadium project, particularly focusing on roof work activities, where workers are most prone to fall hazards. The proposed method aims to provide ease for site workers to access safety knowledge, including risk identification (including risk, fatality, location, and time), visualization of TSFs, personal protective equipment, and safety work instructions. This interactive dashboard also enables safety managers to plan safety measures, allocate TSFs efficiently, and make well-informed decisions to effectively mitigate risks.
Computer vision-based safety monitoring requires machine learning models trained on generalized datasets covering various viewpoints, surface properties, and lighting conditions. However, capturing high-quality and extensive datasets for some construction scenarios is challenging at real job sites due to the risky nature of construction scenarios. Previous methods have proposed synthetic data generation techniques involving 2D background randomization with virtual objects in game-based engines. While there has been extensive work on utilizing 360-degree images for various purposes, no study has yet employed 360-degree images for generating synthetic data specifically tailored for construction sites. To improve the synthetic data generation process, this study proposes a 360-degree images-based synthetic data generation approach using Unity 3D game engine. The approach efficiently generates a sizable dataset with better dimensions and scaling, encompassing a range of camera positions with randomized lighting intensities. To check the effectiveness of our proposed method, we conducted a subjective evaluation, considering three key factors: object positioning, scaling in terms of object respective size, and the overall size of the generated dataset. The synthesized images illustrate the visual improvement in all three factors. By offering an improved data generation method for training safety-focused computer vision models, this research has the potential to significantly enhance the automation of the construction safety monitoring process, and hence, this method can bring substantial benefits to the construction industry by improving operational efficiency and reinforcing safety measures for workers
A significant challenge has long persisted in the construction industry: the lack of a robust incentive system to encourage and motivate workers to prioritize safety. While safety culture has been recognized as crucial, traditional approaches to incentivizing safe behaviours often encounter roadblocks, such as heavy documentation processes, recognition delays, and resource allocation difficulties. This paper addresses this problem by introducing an innovative approach to incentivize and cultivate a safety culture in the construction industry. iSafeincentive integrates blockchain technology and computer vision to develop a novel solution revolutionizing safety monitoring and incentive distribution. Computer vision technology is employed for real-time analysis of safety conditions based on on-site images, ensuring the immediate identification of safe practices. Simultaneously, blockchain technology safeguards the incentive distribution process's integrity and transparency, addressing traditional methods' shortcomings. The findings suggest that iSafeincentive offers an efficient and secure method for rewarding safe activities among workers. Furthermore, the integrated platform offers a promising pathway to enhance job site safety practices, ultimately reducing accidents and incidents within the construction sector
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