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.

• Automated safety tools may exceed manual methods but still need industry adoption. • Decision-making in construction is hindered by limited safety insight extraction. • This study utilizes knowledge graph for enhanced safety and planning insights. • The proposed framework and ontology create effective safety knowledge representation. • A case study demonstrates the cross-domain impact of safety and planning decisions. Construction projects and their dynamic and yet hazardous work environments face significant challenges. Despite advancements, many proposed solutions for information extraction and utilization remain impractical due to complexity and lack of interoperability. Information is often siloed in proprietary formats, making it difficult to integrate. This issue is evident in the construction safety domain, where advanced risk analysis tools provide detailed insights to hazards but can be overwhelming. Similar challenges exist in cost estimation, schedule evaluation, progress monitoring, and quality compliance checking. Decision-making in construction scheduling struggles to assess how changes impact site safety due to insufficient information and knowledge extraction capabilities, especially when it comes to cross-domain knowledge extraction. This study aims to make safety information accessible to safety and planning professionals. By leveraging Digital Twins, automated safety analysis, and knowledge representation, we enable decision-makers to gain deeper insights into their domain and understand the interplay between project planning and safety. We propose a framework for knowledge extraction, an ontology for capturing knowledge, and query building blocks to transform natural language questions into actionable queries. These methods are tested in a case study, revealing valuable insights into the cross-domain impact of decisions.

Specific occupational construction safety, health, and well-being related knowledge and information are scattered and fragmented. Despite technological advancements of information and knowledge management, a link between safety management and information models is still missing. In this paper we present first steps towards a unifying formal (logic-based) domain model of construction safety, called SafeConDM, that consists of: (1) a semantically rich ontology of hazard, safety concepts, and concept relationships that builds on, and integrates with, existing construction safety ontologies and building information models; (2) a set of first-order if-then rules linking construction site states with the potential for specific hazards to occur that we define in a novel way using spatial artefacts. We present a prototype software tool, based on our ASP4BIM tool that implements SafeConDM for construction hazard analysis and safe construction planning decision support, and empirically evaluate our tool on three real-world construction building models.

Civil engineering has only recently started the digitalisation journey by standardising around Building Information Models (BIMs). In the process of construction a dimension of time is added in what is called 4D BIM and this can serve as the basis for a digital twin. It is predicted that such a digital twin can enhance the overall overview of status of the construction of a new building by means of different types of sensors, and interpreting these in relation to a BIM. In the construction phase there are rules and regulations targeting the safety of the different kinds of construction workers at the construction site. In this paper we provide a vision of how digital twins can assist with spotting potential violations of the constraints stated by the rules and regulations, and empirically evaluate a proof-of-concept software tool on a large scale, real-world 4D BIM.

We present a new approach based on spatial reasoning in Answer Set Programming (ASP), and a prototype software tool, for automatically evaluating construction safety compliance of real-world Building Information Models (BIM) that have both a geometric component and temporal component in the form of a construction plan and schedule (4D BIM). In the 4D BIM domain, geometries of building objects are large and complex making it highly impractical to represent geometries as ASP facts, unoptimised spatial reasoning can be prohibitively slow, and rounding errors in floating point arithmetic often result in logical contradictions. Our novel framework addresses these challenges by integrating a specialised geometry database, built-in spatial optimisations, and support for real arithmetic solving. We empirically evaluate our prototype software tool on two large 4D BIM models from real buildings to demonstrate the practicality and scalability of our new framework to real-world workplace hazard prevention tasks in construction safety-in-design analysis.

Time-of-Flight (TOF) sensors provide range information at each pixel in addition to intensity information. They are becoming more widely available and more affordable. This paper examines the utility of dense TOF range data for image segmentation and tracking. Energy based formulations for image segmentation are used, which consist of a data term and a smoothness term. The paper proposes novel methods to incorporate range information, obtained from the TOF sensor, into the data and the smoothness term of the energy. Graph cut is used to minimize the energy.

This paper presents an online, recursive filtering strategy for contour-based tracking. Approaching the tracking problem from an estimation perspective leads to an observer design for the visual track signal associated with an individual target in an image sequence. The track state of the observer is decomposed into group and shape components that describe the gross location and the nonrigid shape, respectively, of the object. A probabilistic representation describes the shape nonparametrically. The constitutive components of the observer are detailed, which include a dynamical prediction model and a correction mechanism. Incorporating the probabilistic observer into the tracking process leads to improved performance and segmentations. The improvements are validated through application of the observer to recorded imagery with evaluation via objective measures of quality.

Online contour-based tracking is considered through the estimation perspective. We propose a recursive dynamic filtering solution to the tracking problem. The state of the target is described by a pose state which represents the ensemble movement and a shape state which represents the local deformations. The shape state of the filter is described implicitly by a probability field with prediction and correction mechanisms expressed accordingly. The filtering procedure decouples the pose and shape estimation. Experiments conducted with objective measures of quality demonstrate improved tracking.