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

Enhancing agricultural aviation safety analysis: a knowledge graph and ontology-based approach to systemic accident pattern detection

Brazilian agricultural aviation combines one of the world’s largest fleets with a rising accident rate, yet existing safety analyses remain predominantly descriptive and single-event in scope. This paper proposes and demonstrates a knowledge graph and ontology-based framework for the systemic investigation of agricultural aviation accidents, using final investigation reports published by the Brazilian Centre for Investigation and Prevention of Aeronautical Accidents (CENIPA) between 2017 and 2024. A domain ontology formalised in OWL 2 (Web Ontology Language) and aligned with the ICAO ADREP taxonomy and the Human–Aircraft–Environment–Management (HAEM) framework was instantiated as a Neo4j property graph over 16 final reports, yielding 67 weighted event–factor relationships across 20 causal factors. Weighted degree, betweenness and closeness centrality, Louvain community detection, and semantic inference rules were applied to identify systemic patterns invisible to retrospective single-event analysis. Community detection converged at modularity Q = 0.207, producing three risk clusters—Powerplant Failure Syndrome, Organisational Deficiency Cascade, and Low-Altitude Control Loss—confirmed by independent qualitative content analysis. Betweenness centrality confirmed that latent conditions—maintenance deficiency and supervisory failure—function as causal bridges, outranking all active failures and validating Reason’s accident causation model at the network level. The network-derived priority order redirects regulatory focus from pilot-centred interventions towards upstream maintenance oversight and operational supervision adequacy. The study delivers the first machine-readable ontology for Brazilian agricultural aviation accident causation and a replicable graph-based analytical pipeline extensible to other specialised aviation segments.

AI-Driven Semantic Processing of Notices to Air Missions for Aeronautical Information Management: Ontology-Grounded Type-Aware Extraction

Timely and accurate interpretation of Notices to Air Missions (NOTAMs) is essential for effective aeronautical information management (AIM). Although NOTAMs follow a standardized format, their abbreviated textual content and heterogeneous operational semantics make automated processing difficult. Existing methods often address type classification, information extraction, and semantic representation independently, leaving a gap between textual interpretation and ontology-grounded information management. To address this gap, this paper presents NOTE, an AI-driven ontology-grounded type-aware extraction framework for NOTAMs. NOTE is supported by a dynamic semantic ontology that combines a shared semantic module with seven type-specific sub-ontologies. Guided by the predicted semantic type, the framework activates the corresponding ontology-aligned schema contract, extracts and validates the relevant information, and materializes the result as Resource Description Framework (RDF) triples. Experiments on 25,341 operational NOTAMs and a manually curated ontology-mapping set produced a macro-F1 of 0.9424 for type-aware routing and an F1 of 0.9053 for structured extraction, with a schema conformance rate of 0.9976. The dynamic ontology achieved slot- and relation-level F1 scores of 0.9472 and 0.9835, respectively. Compared with a scale-matched static ontology, type-specific activation improved relation mapping by 13.06 percentage points while producing semantic richness close to the human-annotated reference. These findings indicate that NOTE provides an effective connection between abbreviated NOTAM text and validated ontology-grounded representations for AIM.

Towards Cyber-Physical Cognition: A Unified Ontology-Driven Knowledge Graph for Real-Time Autonomous Grid Operations

This paper introduces a universal ontology framework for the operational representation of intelligent cyber-physical power systems via a unified knowledge graph and an ontology capable of cross-domain reasoning, and validates the framework's efficacy for real-time decision support.

Semantic Interoperability in Smart Cities: A Distributed Architecture Based on Ontologies, Plug-and-Play AI, and BlockDAG

Smart City platforms increasingly depend on sensing devices and distributed services. The implementation of data-driven mobility systems and infrastructure management has not solved the problem of semantic interoperability. The problem comes from data models in middleware solutions and from the lack of support for devices connecting on their own. Existing approaches often struggle to scale under high data rates and dynamic environments. We presents a work-in-progress distributed architecture that combines semantic technologies, Plug-and-Play Artificial Intelligence (PnP-AI), and a Directed Acyclic Graph (DAG)-based synchronization layer that aims to enable decentralized semantic interoperability in Smart City scenarios. The system applies an Intelligent Transportation System ontology to perform semantic labeling of sensor observation, Shapes Constraint Language (SHACL) shapes for local semantic validation, and a lightweight PnP-AI agent with a Decision Tree classifier to perform automatic device categorization during onboarding. The system design allows nodes to perform processing operations and validation tasks while semantically validated events are propagated through a DAG-based ledger without relying on centralized middleware. Preliminary experimental results from a simulated multi-node environment showed that end-to-end latency remained under 15 ms and throughput reached more than 100 events per second while all messages successfully passed SHACL validation. These results show that the architecture can improve scalability and resilience compared to centralized middleware systems. Current efforts are directed toward expanding the test scope, incorporating security and privacy mechanisms, and demonstrating the approach in more complex urban mobility scenarios.

An Agentic Framework for Dynamic Ontology Generation

Agentic AI-enhanced digital twins for Smart City civil infrastructure: A secure, autonomous and auditable management framework

Smart city implementation increasingly relies on sensing and analytics; however, a persistent operational gap remains between anomaly detection and safe, timely, and accountable intervention in civil infrastructure systems. This paper proposes an Agentic AI-supported Digital Twin framework for smart city civil infrastructure management, where monitoring and action are linked and auditability is maintained. The Digital Twin continuously updates asset and network models of bridges, roads, and water infrastructure using multi-stream telemetry, incorporating state estimation, predictive maintenance, and what-if simulation services. At the orchestration layer, an agent-based Perception–Conceptualization–Action workflow implemented with LangChain and LangGraph enables cross-domain reasoning and coordinated mitigation planning through controlled API calls to municipal data. A permissioned blockchain cryptographically binds observations, approvals, and executed interventions, ensuring provenance, governance, and tamper evidence. To evaluate the framework, 18,000 incident simulations were conducted across five architectural configurations and three scenario complexity levels over 30 independent runs. This simulation study characterises framework behaviour under controlled stochastic conditions and does not constitute real-world operational validation. Ablation analysis isolates each component’s contribution, demonstrating that latency and mitigation gains are primarily attributable to multi-agent orchestration, while the blockchain layer drives decision auditability. Across all configurations, the fully agentic system substantially outperforms the rule-based baseline: mean detection latency of 3,197 s vs. 39,374 s, mitigation success rate of 66.2% vs. 45.5%, blockchain-anchored decision justification of 71.8% vs. 0%, and operator workload reduction of 91.7% vs. 0%. These results demonstrate that combining simulation-enabled digital twins with governance-aware agentic orchestration measurably improves response efficiency, recommendation quality, and action accountability within the bounds of a synthetic evaluation environment.