From smart factories to healthcare: AI redefines the Internet of Things

IoT networks are expanding into nearly every sector of the global economy, but their growing complexity has exposed vulnerabilities in security, energy management and operational control. Traditional rule-based systems are proving insufficient in environments defined by constant data flow and uncertainty. A new scholarly overview highlights artificial intelligence as the driving force behind the next phase of IoT evolution.

The editorial Applications of Artificial Intelligence in the IoT, published in Applied Sciences, brings together this emerging research. The paper introduces a Special Issue showcasing how AI-enabled solutions are transforming industrial systems, smart environments and healthcare delivery.

Securing and optimizing Industrial IoT systems

Industrial control systems depend on the accurate and uninterrupted operation of monitoring, transmission and control subsystems. As these systems become digitized and interconnected, they are increasingly exposed to cyber threats that conventional security tools often fail to detect or mitigate.

A featured contribution explores how large language models can be integrated into industrial IoT security frameworks. The study investigates the use of advanced generative AI systems to support the initial triage of cybersecurity incidents. By leveraging reasoning capabilities and pattern recognition techniques, the proposed approach enhances the classification and prioritization of cyber incidents within operational constraints. This signals a broader shift toward embedding generative AI into industrial defense mechanisms, allowing systems to respond more intelligently to evolving threats.

Energy efficiency represents another major industrial priority. Smart environments now rely on interconnected IoT devices and edge computing networks to monitor and manage energy flows. However, the sheer volume of sensor data can overwhelm traditional analytics systems. Another contribution within the Special Issue provides a comprehensive review of how AI-based data analytics combined with IoT and edge computing can improve energy efficiency.

The review brings together findings from multiple disciplines, including computer science and environmental engineering, to show that AI-enhanced analytics at the edge can extract actionable insights while reducing latency and energy waste. By processing data closer to the source, edge computing reduces reliance on centralized cloud resources, thereby lowering transmission costs and improving overall system sustainability. The integration of AI algorithms allows predictive optimization of energy consumption patterns in smart environments, reinforcing the link between digital transformation and environmental responsibility.

Intelligent supply chains and ecological monitoring

AI-enabled IoT systems are redefining supply chain resilience. Global supply networks face increasing volatility due to environmental uncertainty, fluctuating demand and geopolitical disruption. Traditional decision-making models based on static parameters struggle to adapt to such complexity.

One of the studies highlighted in the Special Issue introduces a fuzzy multi-objective supply chain model that integrates IoT data streams with digital twin architectures and adaptive learning mechanisms. Digital twins create virtual simulations of physical supply chain systems, enabling continuous monitoring and scenario testing. By coupling these simulations with real-time IoT inputs, organizations can adjust operational decisions dynamically.

The model balances multiple objectives simultaneously, including sustainability, agility and robustness. Under environmental uncertainty, the system continuously refines supply chain decisions using adaptive learning algorithms. This demonstrates how AI-enhanced IoT frameworks can support both strategic and operational decision-making, contributing to more resilient and environmentally responsible supply networks.

Environmental and biological monitoring also benefit from advanced AI techniques. IoT systems are increasingly deployed in ecological contexts where accurate identification and classification are essential. A study within the Special Issue applies graph neural networks to IoT-based image acquisition systems for honey bee verification.

By converting captured images into graph representations, the proposed approach improves classification accuracy compared to earlier methods. This innovation highlights the potential of graph-based deep learning models to analyze complex structured data generated by IoT devices. In practical terms, such applications can enhance biodiversity monitoring, agricultural management and ecological research. The work illustrates that AI integration extends beyond industrial automation into broader environmental stewardship.

Transforming healthcare through AI-enabled IoT

Healthcare emerges as one of the most impactful domains for AI-enabled IoT integration. Wearable sensors and ambient IoT devices now generate continuous streams of health-related data, offering unprecedented opportunities for patient monitoring and chronic disease management. However, raw data alone is insufficient. Intelligent interpretation is required to translate sensor signals into meaningful clinical insights.

One contribution in the Special Issue presents an integrated AI-IoT framework designed to analyze patient adherence to treatment protocols. By applying machine learning techniques to sensor-generated data, the system identifies behavioral patterns and detects deviations from prescribed regimens. This enables healthcare providers to intervene proactively and tailor support strategies to individual patients.

The approach underscores the shift toward patient-centered, data-driven healthcare models. Instead of relying solely on periodic clinical visits, AI-enabled IoT systems allow continuous monitoring and adaptive intervention. Such systems can improve long-term outcomes in chronic disease management by enhancing treatment adherence and early detection of risk factors.

Another healthcare-focused study introduces an AI-driven dashboard designed to support frailty detection and monitoring. The system integrates IoT data streams with AI analytics and web technologies to generate intuitive visual insights for caregivers and healthcare professionals. By prioritizing usability and human-centered design, the framework ensures that complex analytics translate into actionable decision support.

The dashboard approach highlights a critical dimension of AI-IoT systems: the importance of effective visualization and user interaction. Advanced analytics must be accessible to end users to deliver real-world impact. The integration of AI with intuitive interfaces bridges the gap between technical innovation and practical healthcare application.