Applications of Artificial Intelligence in the IoT

1. Introduction

Internet of Things (IoT) systems face various challenges in real-world applications, including operational, performance and security issues. As the architecture of sensors, devices, edge networks, transport and core communication systems, and cloud infrastructure are becoming increasingly complex, these management, performance and security challenges become interrelated and difficult to address without AI-based support. The tremendous advancements in Artificial intelligence (AI) have opened new horizons for addressing the dynamic context of today’s industrial environments and tackling their real-world challenges.

The Integration of AI with the IoT ecosystem enables more effective management of performance, operational, and security issues. AI techniques provide intelligent solutions to collect, analyze, and act on data, thereby fulfilling the requirements of real-world applications while supporting sustainability and agility under contextual variability and uncertainty.

This Special Issue presents a broad range of methodological approaches for Industrial and healthcare IoT applications across six high-quality papers. These contributions provide innovative solutions based on advanced AI techniques and address a wide spectrum of real-world industrial and healthcare applications.

2. Protection and Security of Industrial IoT Systems Using LLM

Security and resilience of industrial IoT systems are vital for the critical infrastructure contexts where technological processes are strongly dependent on correct operation of control, monitoring and transmission subsystems. In practice, the data and information processed in these subsystems become the subject of cyber incidents intentionally or accidentally. The complexity of industrial systems exposes them to sophisticated cyber threats that traditional security mechanisms often fail to detect or mitigate. New AI advancements such as LLMs can play a crucial role in this regard.

The paper “Application of Large Language Models in the Protection of Industrial IoT Systems for Critical Infrastructure” by Manowska and Syta [Contribution 1] addresses this challenge by exploring the use of large language models (LLMs) as an intelligent component for industrial IoT security. The study investigates how LLMs can support the initial triage of cybersecurity incidents in Industrial IoT environments. By leveraging the reasoning and pattern-recognition capabilities of LLMs fine trained on an experimental dataset, the proposed approach demonstrated its effectiveness to support incident classification and response prioritization and decision making under operational constraints. This contribution highlights a promising research direction where generative and language-based AI models are integrated into operational IoT security frameworks.

3. AI Powered Data Analytics for Energy-Efficient IoT-Edge

Energy efficiency and sustainability are indispensable in smart environments where IoT systems are integrated into new network infrastructures that include edge networks. AI-powered data analytics plays a vital role in extracting actionable insights from sensor data and optimizing IoT and edge networks. The impacts of AI-based data analytics and edge computing on energy efficiency in smart environments, along with application areas and methods, have been investigated in recent years.

The article “The Impact of Data Analytics Based on Internet of Things, Edge Computing, and Artificial Intelligence on Energy Efficiency in Smart Environment” by Rojek et al. [Contribution 2] presents a comprehensive review on the literature on AI-based data analytics and edge computing for energy efficiency. Using Partial PRISMA as standard, the authors carried out a systematic yet flexible literature mapping, allowing for the identification of scattered research results published in computer science, energy, and environmental engineering. In this paper, the authors investigate how AI-enhanced analytics, combined with edge computing, can improve energy efficiency in smart environments. This work exemplifies how the synergy between IoT, AI, and edge computing contributes to sustainable and efficient system design.

4. Smart Integration of IoT and Digital Twins; Intelligent Supply Chain

Global supply chains are increasingly complex and vulnerable to uncertainty, disruptions, and environmental constraints. IoT technologies enable the collection of real-time data and visibility across supply chain nodes and operations, while digital twins offer virtual simulations of physical systems, facilitating scenario testing, anomaly detection, and predictive analytics. The literature on agile and sustainable supply chains highlights the growing need for systems that can rapidly adapt to disruptions, reduce environmental impacts, and remain resilient in the face of uncertainty. Classical decision-making approaches based on deterministic parameters are increasingly viewed as inadequate in addressing challenges such as fluctuating demand, resource variability, and sustainability requirements, while AI-enabled IoT can address these challenges by using digital twins.

The paper “A Fuzzy Multi-Objective Sustainable and Agile Supply Chain Model Based on Digital Twin and Internet of Things with Adaptive Learning Under Environmental Uncertainty” by Nozari et al. [Contribution 3] introduces a model that leverages IoT and digital twin (DT) architectures to continuously update and refine supply chain decisions in real-time. The proposed technique addresses multiple objectives, including sustainability, agility, and robustness under uncertainty. This work demonstrates how AI-enhanced IoT systems can support strategic and operational decision-making in complex socio-technical systems, contributing to more resilient and sustainable supply chains.

5. Advanced AI Models for IoT; Honey Bee Monitoring Case Study

IoT technologies are increasingly used in biological and environmental monitoring, where accurate identification and classification tasks are critical. Recent advances in deep learning and graph-based models offer powerful tools for analyzing complex, structured data generated by such systems.

The contribution “Apis mellifera Bee Verification with IoT and Graph Neural Network” by Velarde Martínez et al. [Contribution 4] presents an innovative application of graph neural networks (GNNs) for bee verification using IoT-based image acquisition. After converting the captured images to an innovative graph, the recognition by GNN is carried out, where improved classification performance is achieved over the previous techniques. This work, proposing new solutions for the practical challenges in ecological and biological monitoring, illustrates the potential of advanced AI models in IoT-based applications.

6. AI-Enabled IoT in Healthcare

Healthcare represents one of the most impactful application areas for AI-enabled IoT systems, particularly in the context of chronic disease management and patient monitoring. Wearable sensors and ambient IoT devices generate continuous streams of health-related data, which require intelligent interpretation to support clinical decision-making and patient adherence. Another example of IoT applications in healthcare is intelligent dashboard. Beyond data collection and analysis, effective visualization and user interaction are essential for translating AI–IoT insights into practical decision-making. Intelligent dashboards play a key role in presenting complex information in an accessible and actionable form.

In “Understanding Patient Adherence Through Sensor Data: An Integrated Approach to Chronic Disease Management”, Díaz-Jiménez et al. [Contribution 5] propose an integrated AI-IoT framework that analyzes sensor data to assess patient adherence to treatment protocols. The study highlights how machine learning techniques can identify behavioral patterns and deviations, enabling healthcare providers to deliver personalized interventions. This contribution underscores the role of AI-driven IoT systems in supporting proactive, patient-centered healthcare and improving long-term treatment outcomes.

The paper “Frailty Insights Detection System (FIDS)—A Comprehensive and Intuitive Dashboard Using Artificial Intelligence and Web Technologies” by Ciubotaru et al. [Contribution 6] introduces an AI-driven dashboard designed to support frailty detection and monitoring. By integrating IoT data, AI-based analytics, and web technologies, the proposed system provides intuitive visual insights for caregivers and healthcare professionals. This contribution emphasizes the importance of human-centered design in AI-enabled IoT systems and demonstrates how intelligent interfaces enhance usability and real-world impact.

7. Concluding Words

The papers published in this Special Issue highlight the essential role of Artificial Intelligence in enhancing the capabilities, resilience and applicability of IoT systems. From securing industrial IoT and optimizing energy usage to strengthening supply chain protection, enabling intelligent environmental monitoring, and improving healthcare delivery, these contributions highlight how AI-driven IoT solutions are shaping more intelligent, adaptive, and sustainable systems.