Bowtie analysis and LOPA (Layer of Protection Analysis) are both risk assessment methods, but they serve different purposes. Bowtie analysis provides a visual representation of risks and the barriers in place to prevent or mitigate them, while LOPA serves as a semi-quantitative method that focuses on evaluating the effectiveness of specific safety layers (barriers) in reducing risk. Essentially, LOPA enhances bowtie analysis by adding quantitative data to assess the effectiveness of the barriers identified in the bowtie diagram. And yet, it is assumed that significant improvements can be made in construction risk assessment once smart technology is applied to existing risk assessment practices. This paper introduces the state-of-the-art practice in the construction sector and critically reviews existing challenges and future opportunities for technology to guide as an additional layer protection and enhance the safety performance in construction companies. Key findings formulate a forward-looking pro-active risk assessment framework that, once validated, help decision makers to successfully integrate technology in existing construction safety management strategies.

Tower cranes are critical for lifting heavy elements in prefabricated and modular construction, but their operations pose significant safety risks. This study proposes a data-driven method using Real-Time Kinematic Global Navigation Satellite System (RTK-GNSS) that tracks the crane trolley and workers movements and assesses the potential struck-by hazards from lifted payloads. By analyzing the trolley’s velocity and vertical displacement, the novelty in the method respectively detects the lifting stages and estimates the payload weight, while pairing its real to the planned placement location in the Industry Foundation Classes (IFC) model. An energy-based hazard assessment computes the intensity of detected incidents when workers’ RTK-GNSS wearables are inside of hazardous crane swing zones. The severity is evaluated in the form of a statistical analysis and density map. Unprecedented safety-relevant information becomes available to practitioners that can use it in responsible decision-making. Compared to computationally intensive systems such as cameras, the proposed method provides an alternative cost-effective, scalable solution for automating the monitoring of hazardous outdoor workspaces.

Construction safety risk assessments are labor-intensive and disconnected from dynamic project conditions and digital systems. This research investigates how domain-specific multi-agent generative artificial intelligence systems can automate construction safety risk assessments and subsequently share them with knowledge graphs (KGs) to enable downstream reuse of the extracted knowledge and overcome information isolation. The research employs a Design Science Research methodology to develop a novel multi-agent system leveraging Large Language Models (LLMs) and semantic-web techniques. Individual agents within the system perform specific tasks, including hazard identification, risk mitigation, and ontology-aligned SPARQL query production. An evaluation with eight construction safety experts validates that the system successfully automates the production of complete and relevant risk assessment documents. A semantic, structural, and syntactical correctness assessment through SHACL shapes, and manual completeness evaluation, furthermore validates that the system reliably converts the risk information into accurate SPARQL insertions compatible with existing ontology-based KGs. Unlike prior research, this agentic pipeline automates task-based assessment generation and semantic integration, enabling downstream reasoning and interoperability within digital twin environments. Safety professionals benefit from faster and machine-readable availability of safety information that supports downstream reasoning processes (e.g., hazard detection or simulation), which are not evaluated in this study. Future research should identify whether the provided knowledge is sufficient to streamline such downstream processes and outline clear information requirements to enhance the system further. Future research should investigate automated and scalable semantic validation methods, reasoning over concurrent tasks, and integration with digital twin systems.

Due to the dynamic nature of construction sites, constant installation and removal of safety equipment is a required practice. This to date requires manual, thus infrequent and still time-consuming inspections to ensure the safety measures are correctly in place when needed. This paper introduces a novel end-to-end pipeline that integrates SafeConAI, UAV-collected point cloud data, and 4D BIM to automate safety inspections, overcoming the limitations of fragmented prior approaches by enabling real-time deviation mapping and comprehensive safety monitoring. This approach aims to bridge the gap between as-built data integration with autonomous navigation by providing location-aware compliance checking. Autonomous ground or aerial vehicles collect point clouds, and a segmentation model is trained to detect and segment safety guardrails and other essential building elements. The system is built and tested in a laboratory environment first by creating a typical fall-from-height protection case. The information generated from the point cloud segmentation is then compared with the original BIM to monitor and point out deviations, and finally provides an analysis report in a user-friendly SafeBIM. The method has been evaluated on several test cases from laboratory to real construction site settings. The results from this work indicate a promising method that can assist practitioners in (semi-) automating construction site safety inspection and making workplaces safer.

Construction automation remains fragmented, with Building Information Modeling (BIM), Additive Manufacturing (AM), and robotics operating as isolated technologies rather than integrated systems. This research presents an early-stage framework that explores the potential for connecting digital design to physical assembly through Industry Foundation Classes (IFC) data exchange within a Construction Learning Factory (CLF) environment. Implemented at 1:50 scale, the framework demonstrates the feasibility of extracting data for both manufacturing-ready geometries (STL) and robot navigation (IFC-derived coordinates) from a single BIM model to coordinate separate automation processes. Seven high-detail modular building components were successfully fabricated and assembled, validating the underlying premise that standardized BIM data can drive both AM and robotic assembly without requiring proprietary data formats or extensive custom programming. The robotic assembly system achieved reliable performance with millimeter positioning accuracy and consistent safety protocols when worksite hazards were simulated, while the manufacturing procedure successfully fabricated detailed modular components that preserved complex building systems including Mechanical, Engineering, and Plumbing (MEP) integration. Although significant development remains to achieve full process integration, this work establishes foundational capabilities for standardized data workflows in construction automation while also providing an educational platform for hands-on exploration of Industry 4.0 technologies. The framework serves as both proof-of-concept validation for IFC-driven automation approaches and a practical foundation for human workforce training inside a CLF environment.

Construction automation remains fragmented, with Building Information Modeling (BIM), Additive Manufacturing (AM), and robotics operating as isolated technologies rather than integrated systems. This research presents and evaluates an integrated framework that connects and automates design, manufacturing, transportation and assembly through Industry Foundation Classes (IFC) data exchange within a physical Construction Learning Factory (CLF). Implemented at 1:50 scale, the framework demonstrates how standardized BIM data can drive coordinated automation across separate processes. Six high-detail modular building components were fabricated through an automated slicing and digital fabrication workflow, transported via autonomous navigation, and assembled in a controlled CLF setup within millimeter-level positioning accuracy. Performance evaluation comparing automated operation against manual control demonstrated measurable differences in consistency, accuracy, and throughput, while revealing inherent trade-offs between speed and reliability, especially in the context of transportation. The complete digital fabrication workflow – from IFC model through automated manufacturing, vision-triggered transport, to coordinate-driven assembly – operates without proprietary middleware, using open-format data exchange and relatively lightweight integration scripts. This work illustrates foundational capabilities for standardized, interoperable data workflows in construction automation. These implemented at larger scale can assist practitioners in real-world construction processes.

Effective knowledge management in construction safety is essential yet challenging. Despite emerging technologies to collect valuable data automatically, it continues to rely on manual input. The heterogeneity of data sources in construction makes it additionally difficult, resulting in a high number of incidents due to late changes in the design. Presented is a unified ontology for construction safety named UNOCS that shares safety knowledge between stakeholders during the construction processes. The UNOCS ontology follows the Linked Open Terms methodology and integrates established concepts, ensuring interoperability with other domain-specific knowledge for multiple use cases: (1) hazard and mitigation planning, (2) conformance checking and control, and (3) incident logging. UNOCS was evaluated through automatic consistency checks, criteria-based assessment, and task-based evaluation. The ontology meets the defined requirements and represents safety-related concepts. Implemented in a machine-readable format, it enables reasoning and seamless knowledge transfer between mitigation planning, safety inspections, and incident reporting.

Construction sites represent hazardous work environments, necessitating comprehensive risk assessments to ensure worker safety. Traditional methods of risk assessment are critical for identifying and mitigating potential hazards. However, the complex nature of these safety documents often poses a significant challenge. This gap in safety knowledge can lead to inadequate risk assessments and heightened workplace hazards. The emergence of advanced technologies in Artificial Intelligence (AI) such as Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) presents new opportunities to address these challenges. While LLMs enhance natural language processing, this hallucination and lack of reliable source referencing can undermine their effectiveness. RAG offers a promising solution by integrating information retrieval, though its application in construction safety engineering remains unexplored. This paper introduces a safety risk assessment platform leveraging agentic AI systems, LLMs, hybrid semantic search, and RAG to generate expert-level safety risk assessments. The system utilized extensive safety knowledge sources from Safe Work Australia into a framework where users can create risk assessment reports for specific work tasks purely based on natural language input. Preliminary evaluations of the platform demonstrate its effectiveness in generating such reports, although improvements are necessary to reduce document verbosity and enhance usability.

Poor construction worksite conditions cause many Serious Injuries and Fatalities (SIF). This paper introduces the Right-time Analysis and Mitigation (RAM) module as part of the Digital Twin for Construction Safety (DTCS) concept. RAM (a) leverages safety rule-based compliance checking of construction layout plans in Building Information Modeling (called safeBIM) and (b) records and analyzes resource trajectory data involved in potential struck-by incidents using wearable and mobile Real-Time Kinematic Global Navigation Satellite Systems (RTKGNSS) technology. RAM further (c) provides new access to unprecedented information and gives detailed insights into human-machine interactions that (d) permit rapid implementation of change by targeted control. Results from real-life validations create an in-depth understanding of the potential root causes that often lead to harmful struck-by incidents. While the proposed DTCS technology contributes data input and information output, the change in safety management processes leads to safer work practices. Both prevent SIF that plague the construction sector.

Construction sites naturally harbor many dangers that expose workers to the risk of injury. These include possible falls, electric shocks, falling objects and the improper use of equipment. To prevent work-related injuries, safety authorities around the world prescribe standards to protect construction workers. The prescribed protective measures include the provision of certified protective equipment, the installation of guardrails and fall protection systems, the proper use of ladders and scaffolding and the implementation of comprehensive training programs. This paper will explore the concept of utilizing a 2.5D platformer serious game for construction safety training. In particular, the conception and implementation will apply design guidelines from game development to realize the relevant content in a promising form, both in terms of motivation and learning success.

Construction sites are among the most dangerous workplaces due to their complex, dynamic, continuously changing work environment. Many existing workplace safety planning techniques rely on two-dimensional drawings and manual expertise. Such efforts are cumbersome as safety plans quickly become outdated as construction work progresses. There has been significant research into automated safety planning, yet no community-wide standard exists for objectively measuring and comparing automated safety assessment efficacy. To address this, an automated performance assessment framework is proposed. It evaluates input solutions regarding newly formalized quantitative soundness, completeness, and spatial correctness indicators. The ground truth of the deadliest hazard falls-from-height is collected through a workshop with domain experts. We validate the proposed framework in a case study, where the performance of our previously developed automated safety planning algorithm is assessed by our new performance assessment framework. The results yield valuable insights into the importance of automated evaluation frameworks that can convince practitioners to invest in human-assisted Prevention through Design and Planning strategies.