GenAI-Driven Occupational Safety Analysis in IoT-Enabled Environments

Industrial Tokenization for LLM-Based Health Intelligence: A Federated Architecture for Industrial Evidence Integration

Industrial health management increasingly relies on heterogeneous information sources, including condition monitoring systems, supervisory control and data acquisition systems, maintenance records, inspection results, and prognostic models. Although large language models provide new opportunities for cross-source reasoning, industrial data and analytical outputs differ substantially in structure, temporal resolution, physical meaning, and reliability. Directly integrating such heterogeneous information into a monolithic model may reduce interpretability, traceability, and adaptability to equipment and data changes. This paper introduces Industrial Tokenization, a conceptual interface for transforming source-specific analytical outputs into structured and machine-interpretable units of industrial evidence, termed Industrial Tokens. Unlike numerical tokens used to encode raw time-series data, Industrial Tokens represent domain-grounded evidence together with source, temporal scope, operating context, analytical meaning, quality or confidence information, and provenance. Based on this concept, a federated industrial architecture is proposed, where heterogeneous analytical subsystems retain autonomy while exposing standardized Industrial Tokens to a central reasoning layer. As an initial implementation, this study presents an end-to-end DiagnosisToken pathway based on vibration-diagnostic outputs, rule-based event aggregation, structured textual token generation, and LLM-based interpretation. Other Industrial Tokens, including SCADA-based condition-monitoring tokens, maintenance tokens, and prognostic tokens, are reserved as future extensions. The proposed framework positions Industrial Tokenization as a semantic interface between domain-specific industrial intelligence and LLM- or agent-based reasoning, rather than another method for encoding raw industrial data.

Navigating Development of Cyber-Physical Systems through Inconsistency Scenarios

An Ontology-Based Human Digital Twin Framework for Proactive Shipboard Safety Intelligence

Maritime operations entail complex interactions between human operators, vessel systems, and dynamic environmental conditions, rendering shipboard safety management a formidable and persistent challenge. Despite advances in automation and monitoring technologies, severe onboard accidents—particularly those related to confined space entry, work at height, hazardous environments, and human error—continue to occur, while existing safety systems remain largely reactive. This paper presents an ongoing study on an ontology-based collaborative shipboard safety analysis framework that integrates artificial intelligence, Human Digital Twin (HDT) modelling, and digital twin–based visualization to support proactive and explainable safety intelligence. The framework is designed to acquire high-density onboard data through multi-source wearable, environmental, spatial, and operational sensors, and to formalize maritime safety regulations and human–environment interaction knowledge into an ontology-driven knowledge base for context-aware risk inference.A multi-layered system architecture encompassing data acquisition, edge-based processing, HDT modelling, AI-driven risk analysis, digital twin simulation, and feedback-driven learning is introduced. This study establishes a foundational architectural and methodological framework for next-generation shipboard safety intelligence and provides a basis for future experimental validation and real-world deployment.

A Fully Automated, Deployment-Aware Testing Pipeline for IoT-Based Automotive Applications

This work presents an end-to-end, deployment-aware testing pipeline for IoT-based automotive applications that combines requirement-driven test and code generation with large language model (LLM) and vision-language model (VLM) assistance, and human-in-the-loop curation to reduce manual effort and improve consistency.

Natural Language-Driven Analytics for Networked Photovoltaic Systems Using LLM-Based Agents

The analysis of photovoltaic (PV) degradation data presents significant challenges due to the complexity and volume of heterogeneous information generated by modern monitoring systems. In contemporary deployments, PV installations operate as distributed, networked cyber-physical systems, where multiple sensing devices and monitoring nodes continuously generate multi-modal data streams. This paper proposes a natural language-driven approach based on Large Language Model (LLM) agents to enhance the accessibility of analytics in such networked environments. We design an agent-based architecture that translates natural language intents into executable analytical workflows, enabling intuitive interaction with data generated by IoT-enabled PV monitoring infrastructures. The system integrates LLM reasoning with structured data processing, visualization tools, and domain-specific context. The proposed approach is implemented using Python and LangChain, and evaluated on the PV EL HDR BW Test DB dataset. Experimental results demonstrate an overall success rate of 83.3% across representative analytical tasks, highlighting the practical viability of LLM-based agents for domain-specific data analysis. This work contributes to AI-driven analytics for distributed and networked systems, showing how natural language interfaces can support data-driven decision-making in emerging intelligent energy and IoT infrastructures.

BIM-Integrated Explainable Vision for Real-Time PPE Safety in Yemen’s Construction Sector

This paper develops a BIM-integrated, explainable, and deployment-aware decision-support framework for real-time personal protective equipment (PPE) monitoring in Yemen’s construction sector. Its contribution is an integration-based decision-support artifact rather than a new PPE detection algorithm. The study addresses a persistent practical gap: many vision-based PPE systems can detect violations, yet they rarely convert image-level detections into auditable, location-aware, and managerially defensible interventions. Using a design science research approach, the paper re-specifies the original detection-centered concept as a socio-technical artifact composed of six tightly coupled layers: multimodal site capture, RF-DETR-based PPE perception, explanation generation, BIM spatial anchoring, AHP-TOPSIS-driven prioritization, and governance-oriented analytics. The assessment remains analytical, and field validation is left for future work. The framework formalizes an event schema that links each alert to confidence, explanation evidence, anchor confidence, zone semantics, response ownership, and closure status. It also introduces a resource-aware deployment path suited to fragile and connectivity-constrained projects by combining smartphone inspections, CCTV streams, offline buffering, staged BIM anchoring, and selective explanation triggering. The main contribution is therefore not merely improved PPE recognition; rather, it is the conversion of real-time vision outputs into trustworthy safety intelligence that supports prioritization, hotspot discovery, accountability, and progressive digital-twin readiness.