Jul 2026· Georgian Maritime Scientific Journal· Vol 3, pp. 22-40· 0 citations· 24 references
TL;DR
This paper explores the integration of Generative AI and Explainable AI into Predictive Maintenance (PdM) within the Shipping 4.0 framework for educational use and proposes a conceptual framework called IPMDP (Intelligent Predictive Maintenance Decision Process for Education), which involves an architectural approach for integrating LLMs (Gen AI approach) into the predictive maintenance of marine engines and systems.
Abstract
This paper explores the integration of Generative AI (Gen AI) and Explainable AI (XAI) into Predictive Maintenance (PdM) within the Shipping 4.0 framework for educational use. The authors propose a conceptual framework called IPMDP (Intelligent Predictive Maintenance Decision Process for Education), which involves an architectural approach (System-of-System, SoS) for integrating LLMs (Gen AI approach) into the predictive maintenance of marine engines and systems, in order to support decision-making in the maintenance of systems at sea. The central idea of the conceptual framework lies in the utilization of a decision-making model using Gen AI techniques through the ExtendAI framework, which functions as a “cognitive partner” to a ship’s engine room crew. The ultimate goal is the educational use of such a framework in the training of Merchant Marine engineers and other ship engine room personnel. Essentially, this study is a conceptual framework paper, integrating existing models and literature to propose a novel application domain, rather than presenting original empirical findings. This cognitive maintenance approach improves operational safety and ensures human-centric, accountable decision-making, which is crucial for regulatory compliance and effective adoption in the maritime industry.
By foregrounding decision intelligence in complex systems, Enactive AI expands the frontier of AI from model capability to system-aware action, opening new possibilities for scalable, governable, and socially valuable AI deployment.
Zuo-Jun Max Shen, Yuan Qu, Pu-Jun Zhang et al.· 0 citations
Abstract Background/Purpose The integration of Robotic Process Automation (RPA) with Enterprise Resource Planning (ERP) systems presents a complex evolutionary challenge. Existing frameworks fail to capture the divergent strategic paths organizations take. This article proposes a three-dimensional conceptual model to explain this phenomenon, defined by the axes of Integration Locus (Resilience), Task Intelligence, and Scale of Automation. Methods A structured literature review of pertinent articles published up to August 1, 2025, was conducted. The findings were used not to generate, but to provide empirical illustration and support for the pre-defined components of the proposed 3D model. A foundational set of 42 papers was identified, from which a curated evidence base of 26 exemplars was selected based on theoretical relevance. Results The literature provides strong evidence for all three dimensions. The evolution along the Resilience Axis (from brittle GUI to robust API) is well-documented. The Intelligence Axis is evidenced by the shift from rule-based bots to cognitive agents. The Scale Axis is shown in the move from discrete task automation to end-to-end process orchestration. Conclusion The 3D model is a robust framework for explaining the trade-offs in automation strategy. It contributes to theory by synthesizing multiple concepts into a new, comprehensive model and provides a diagnostic tool for practice.
The ORBIT (Operations Responses and Business Intelligence Toolkit) project was initiated to assess agentic AI for the upcoming ESnet 7 initiative and to address persistent operational pain points in the Network Operations Center (NOC) workflow. ESnet operators experience slow retrieval from siloed data sources, incidents described in lengthy and difficult-to-parse tickets, and context loss across shift handoffs. These challenges increase cognitive load and prolong incident resolution times. ORBIT therefore targets routine automation, cross-source synthesis, and actionable insights delivered directly within operators'existing tooling. ORBIT is an agentic AI system integrated into ServiceNow, ESnet's primary incident management platform. The design uses a modular, layered architecture comprising a centralized reasoning hub, tool access via MCPs for ESnet data sources, a semantic search layer, and an operator-facing chat interface. To manage the complexity and stochasticity of the AI toolchain, ORBIT follows industry best practices by structuring task logic as versioned, tested"skills"that guide the system in performing bounded responsibilities. This improves reliability and predictability compared to fully unconstrained agent behavior. Key results show that ORBIT successfully delivered all six initial tasks, and the architecture enabled rapid development of two additional tasks proposed by NOC engineers. We observed strong organic adoption of general-purpose infrastructure components, especially the chat interface and LiteLLM model gateway, including high request volumes from outside the project. Experiments with skills indicate that this approach can reduce task completion steps while eliminating observed error modes.
BinHao Dong, Sukhada Gholba, Brooklin Gore et al.· arXiv.org· 0 citations
Decision-making latency and high cognitive load often hamper hotel operations due to the passive design of modern Property Management Systems (PMS). Although the adoption of artificial intelligence (AI) in the hospitality sector is increasing, its implementation is still focused on the guest-facing services. Conversely, in the back office, most inefficiencies remain unresolved. This paper addresses this gap by developing an agentic AI co-pilot designed to assist front office staff as a proactive operational partner using natural language interaction. Using the ReAct (Reasoning and Acting) framework, static and administrative workflows are reorganized into an autonomous execution loop. The system is designed by integrating a Large Language Model (LLM) with a Server-Side Rendering (SSR) environment. Comparative analysis evaluated efficiency, accuracy, and robustness. Results show that the AI agent performs 6.34 times faster in generating performance reports. For the daily reservation workflow, there is also a 2.32-fold increase in speed compared to interactions through passive GUI navigation. Regarding reliability, the standard LLM had a hallucination rate of 76.7%, while the proposed agentic AI co-pilot demonstrated significantly higher reliability and achieved 90% resilience against robustness testing. Although the agent encountered minor failures, it achieved an average success rate of 93.3% across 30 test scenarios. Overall, the AI co-pilot with the ReAct framework has shown significant efficacy in addressing hotel operational inefficiencies.
Bernhardth Neo Videan, Rio Nurtantyana· International Conference on...· 0 citations
Sports analytics platforms have come a long way in their ability to handle data themselves and their performance in visualization, but lags remain when it comes to human-centric decision support, explainability, and contextual reasoning. This study aims to map out the Contextual Decision Intelligence for Sports Analytics (CoDI-SA) framework to a feasible reference architecture, based on the Design Science Research (DSR) method. The proposed architecture includes multi-modal data fusion, contextual reasoning, explainable artificial intelligence (XAI), and human-in-the-loop decision intelligence, which will allow for intuitive, context-aware and actionable suggestions to coaches, analysts and performance teams while keeping transparency. The proposed architecture is adaptive in nature and can be implemented on elite level, semi-professional level or grassroots level, all while having a common architectural foundation and is an alternative to the existing commercial systems which emphasize descriptive analytics. The usefulness and trustworthiness of the architecture as well as the value of the architecture is assessed by comparing it to commercial sports analytics platforms and by surveying twelve sports-technology practitioners with a grand mean of 4.26/5 and SD of 0.69 for the three questions, where the mean represents the average of the practitioners' responses. This study offers a reference architecture that can be reused, and seven design principles and a tiered deployment model of next generation sports analytics systems.
S. Sangle· 2026 International Conferenc...· 0 citations
The current research paper proposes an AI-driven framework for optimizing processes of mega infrastructures within the framework of the Saudi Vision 2030 program. The authors are convinced that AI technology should not be perceived as something separate from other phenomena; on the contrary, it is the layer of orchestration allowing linking such components as planning information, BIM models, digital twin infrastructures, IoT technologies, logistics processes, data, and decision-making processes at the executive level in the feedback loop. The present paper provides a systematic review of the most recent studies in the field of applying AI technology in construction management, BIM intelligent systems, digital twin infrastructure, engineering with digital modeling and simulation, environmental analyses for construction and operation, and the transition to the Digital Saudi Arabia which have been published between 2020 and 2025. The possibility of applying machine learning, computer vision, generative optimization, predictive control, and dashboard intelligence to enhance the reliability of the schedule, efficiency in material use, predictability of finances, quality assurance and control, and risk transparency during the management of large-scale projects is studied. The present research is concentrated on identifying the process architecture including such components as data model standardization, stage gate decision and management, interoperability, notifications and explanations, and human intervention which are essential for implementing the AI technology and achieving its maximum advantages. As a result of the research, the authors propose an AI-Driven Process Optimization Framework for Saudi Mega Projects (AIPOF-SMP).
Syed Mohammad Danish Jafri· Veredas do Direito· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
