Search Results (653)

Background: Internet of Things (IoT) deployments face complex network dependencies and persistent data limitations that constrain machine learning (ML) intrusion detection systems (IDS). Objective: To synthesise peer-reviewed machine learning IDS research for IoT, focusing on dependency-driven failure propagation and chronic data scarcity, positioned against 2024–2025 EU regulatory requirements (NIS2, the Cyber Resilience Act). Eligibility criteria: Peer-reviewed empirical studies proposing or evaluating a machine learning or deep learning IoT intrusion detection method published in English from January 2018 (limited pre-2018 exceptions for seminal works). Information sources: IEEE Xplore, SpringerLink, Elsevier ScienceDirect, Scopus, Web of Science, and Google Scholar, searched on 12 February 2025. Risk of bias: Each candidate was scored against four criteria (objectives clarity, methodological soundness, reproducibility, IoT-security relevance); studies scoring at least 3 out of 4 were retained. Screening and scoring were performed by one reviewer, with a second reviewer independently checking 20 percent of records. Synthesis methods: Narrative thematic synthesis; heterogeneous metrics and incompatible datasets across studies precluded quantitative meta-analysis. Included studies: Of 427 records identified, 52 studies initially met inclusion criteria; a post hoc independently validated reconstruction of individual QA1–QA4 scores subsequently found that six did not meet the threshold or topical eligibility criteria, yielding a final 46-study corpus. Main findings: Generative adversarial networks (GANs) dominate dataset augmentation work, graph-based communication analysis addresses dependency modelling, and methods based on transformers or federated learning emerge from 2023 onward. Certainty of evidence: No formal grading (GRADE) applies to this narrative synthesis. Confidence in the corpus composition is high, following independent QA1–QA4 validation, while confidence in the thematic findings is moderate given single-reviewer screening and judgment-based classification. Conclusions: We identify three recurring gaps: real-time detection under resource constraints, dependency-aware detection, and regulatory compliance. Closing these gaps requires detection methods that treat IoT security as a networked and regulated system rather than an isolated device classification problem. Registration: Open Science Framework, 10.17605/OSF.IO/NMAK4 (registered retrospectively). No external funding supported this review. Full article

The increase in the number and complexity of interconnected systems requires new methods to identify potential threats in today’s hyperconnected world. This trend affects systems ranging from smart homes and Internet of Things (IoT) devices to critical infrastructure which must be equipped with the corresponding cyber defense methods. Given the new landscape, it is more difficult for classical cybersecurity systems to stay up to date with novel threats, as well as to keep track of all interconnected devices defined by various protocols and behaviors. Artificial intelligence (AI) represents a strong candidate to complement traditional cyber defense methods due to its adaptability to variation and capability to identify complex data patterns, which has led researchers to develop state-of-the-art anomaly detection systems. The current critical review aims to analyze the scientific literature on three dimensions including used algorithms and datasets, domain challenges hindering AI deployment in productive environments, and the capability of explainable artificial intelligence (XAI) to support cyber security experts with insights into the model’s inner workings and decision rationale. Compared to existing scientific reviews, this paper moves beyond algorithmic comparison by providing a methodological interpretation of AI anomaly detection landscape, demonstrating how data availability, learning paradigms, and explainability collectively influence the evolution of cyber defense research towards operational deployment. This approach revealed that AI development for cyber defense is highly heterogenous, and that the available datasets strongly influence the algorithm of choice, rather than the models being chosen methodologically based on proven performance. The analysis further indicates that operational deployment remains challenging, as the literature continues to report substantial limitations related to data quality, computational requirements, and model interpretability. Full article

Energy-transition policy increasingly asks households and small firms to behave as informed players in liberalised electricity markets. How far they can actually play that part, however, depends on two things at once: how energy-literate they are, and how clearly their bills speak to them. Although both strands are separately well researched, they are rarely measured jointly, on the same respondents, and at the level of the bill itself—the document through which almost every consumer actually meets the market. Working from a case study of Polish electricity customers, we examine how self-assessed knowledge of billing relates to perceived bill comprehensibility and to the detailed understanding of individual invoice elements. The evidence comes from a mixed-mode CATI/CAWI survey of 602 respondents—individual consumers, prosumers and micro/small enterprises—all served by a single major supplier in the Upper Silesia region. Because the achieved sample is dominated by older respondents of one supplier, the study is framed throughout as an exploratory, regional investigation. Three findings stand out. First, self-assessed energy literacy, perceived bill comprehensibility and the detailed understanding of individual invoice elements are positively interrelated (Spearman’s ρ between 0.49 and 0.61, p < 0.001), indicating that consumer competence and information design are complementary correlates of comprehension. Second, declared knowledge and understanding are socially patterned: they run higher among respondents aged 35–54 and those in a better financial position and lower among older and financially vulnerable customers, while a third of respondents find their current bill hard to understand. Third, Ward’s cluster analysis distinguishes three internally consistent layers of billing information—detailed billing parameters, transactional and payment content, and regulatory and legal content—of which the regulatory and legal layer is read least often. Because the data come from one supplier in one region, the findings are indicative rather than nationally representative and should be generalised only with care. The paper’s contribution is to measure energy literacy, perceived transparency and detailed bill comprehension jointly on the same respondents and to document—for a Central European retail market undergoing rapid price change—that the regulatory layer of the bill is precisely the layer consumers read least. For policy, the findings support piloting simpler bill structures, communication aimed at vulnerable groups, and energy education that equips consumers to take part in the market as the transition unfolds; given the sample, these implications are advanced as directions to be tested rather than as validated prescriptions. Full article

This paper presents a comprehensive review of the integration of Intelligent Transportation Systems (ITS) and Digital Twin (DT) technologies for intelligent traffic management. It examines the role of key enabling technologies, including the Internet of Things (IoT), machine learning (ML), deep learning (DL), reinforcement learning (RL), Graph Neural Networks (GNNs), vehicle-to-everything (V2X) communication, and edge computing. These technologies support real-time traffic monitoring, traffic prediction, and adaptive control in ITS. The review synthesizes recent research on conventional traffic control methods, optimization-based approaches, learning-based techniques, and DT-enabled traffic management solutions. Particular attention is given to the integration of DTs with intelligent traffic signal control, real-time synchronization, multi-intersection coordination, communication latency, sensing uncertainty, and scalability. The reviewed literature demonstrates the potential of DT-enabled ITS to improve traffic efficiency, reduce congestion, enhance transportation safety, and support sustainable mobility through data-driven decision-making. However, significant challenges remain regarding communication delays, sensor and data uncertainty, computational complexity, scalability, and validation under realistic urban conditions. Based on the reviewed literature, this paper identifies key research gaps and outlines future research directions toward scalable, reliable, adaptive, and real-time DT-enabled ITS architectures for next-generation smart cities. Full article

With the acceleration of global urbanization and the large-scale development of underground spaces, underground engineering faces extremely complex and variable geological environments and high-risk construction disturbances. The widespread application of the Internet of Things and novel sensing technologies has given rise to structural health monitoring data increasingly characterized by massive volume, high dimensionality, multi-source heterogeneity, and strong spatiotemporal coupling. Traditional data processing methods based on mechanical analysis, empirical formulas, or numerical simulation have increasingly exposed limitations of insufficient accuracy, lengthy computation times, and weak generalization capability when confronted with such engineering big data. Machine learning and deep learning technologies, by virtue of their superior nonlinear mapping capability, advantages in feature extraction from massive data, and flexible architectural design, provide solutions for efficient knowledge extraction and intelligent assessment of underground engineering monitoring data. This paper reviews the current application status and frontier advances of machine learning technologies in the field of underground engineering monitoring data processing in recent years. First, the development trajectory of analytical algorithms evolving from classical shallow machine learning, through temporal and spatial deep learning, to physics-data dual-driven approaches is delineated. Second, targeting the critical challenges of missing field data and sparse sensor deployment, spatiotemporal fusion imputation techniques and spatial reconstruction methods incorporating mechanical prior knowledge are thoroughly evaluated, elucidating the paradigm shift in monitoring philosophy from discrete point-based alarming to inference-augmented sparse sensing that approximates full-field state awareness through model-dependent estimation rather than direct measurement. Third, the applications of machine learning in underground structural deformation mechanism interpretation, key influencing factor identification based on explainable artificial intelligence (AI), and rapid back-analysis of geomechanical parameters are summarized. Finally, composite network architectures and physics-constrained guidance strategies for non-stationary deformation time series prediction under complex and variable working conditions are discussed. A methodological audit of the 73 included studies—of which 33 enter the quantitative comparison tables—reveals that 26 of the 33 audited studies (78.8%) validate exclusively on single-project data, only 1 study conducts rigorous out-of-distribution generalization testing, and none of the 33 studies (0%) provides uncertainty quantification. These findings highlight cross-project generalization and probabilistic prediction as important methodological challenges. This paper aims to provide theoretical references and methodological guidance for safety early warning, intelligent construction, and full life-cycle health management of underground engineering. Full article

The goal of the study is to compare the most popular protocols available in IoT (Internet of Things) nodes and how they perform across multiple samples in controlled conditions. The focus is on local telemetry transmission, in which the ESP32-C6 and ESP32-S3 boards communicate with two other devices—a router and a laptop—over the 2.4 GHz band. For the experiment, three standard nominal transmission sizes of 16, 64, and 256 bytes were pre-set. Each size was trialed for 30 samples for each protocol. All transmissions achieved a 100% success rate and were received by the end device. Both latency and success rate were used as the main performance indicators. The end results were that CoAP had the lowest mean latency of 47.0 ms, followed by MQTT with 63.0 ms and HTTP with a mean overall latency of 1419.0 ms. A comparative test with the ESP32-S3 revealed that while the overall protocol ranking remained identical, the gap eventually narrowed. The more powerful ESP32-S3 processed HTTP significantly faster (mean latency of 449.0 ms) but yielded slower latencies for the lightweight CoAP and MQTT protocols compared to the C6. The results indicate an order of magnitude faster performance of CoAP and MQTT than HTTP in the test setup. The study offers a simple and reproducible setup and benchmarking, allowing it to serve as a practical reference point when selecting the right protocol for a given use case. Full article

Among the key computing challenges today are green computing and quantum computing, with the quantum paradigm introducing adverse consequences for many existing cryptographic protocols. However, the entire internet critically depends on these protocols. Therefore, building on theoretical preliminaries, this paper derives a coherent methodological basis for designing and developing quantum-computing-resistant and sustainable protocols for sensors and the Internet of Things (IoT) by extending traditional approaches and including large language models (LLMs). In line with these premises, this paper focuses on authenticated key exchange (AKE), which is one key security service. To support ever-growing needs under the described conditions and using the developed methodology (called the Sustainability-focused LLM-supported crypto protocols development process, SFLSCDP), this paper presents a novel, formally verified AKE protocol family. This family of protocols can be considered ultra-lightweight, as a corresponding metric is also given. Consequently, additional effective means for security services in the post-quantum computing era are provided, even for computationally weak devices. Full article

Digital Twin (DT) technology has emerged as a transformative approach in pipeline engineering, enabling real-time monitoring, predictive analytics, and enhanced decision-making across the asset lifecycle. This review critically examines recent advancements in the application of digital twins for pipeline systems, with a particular focus on condition monitoring, leak detection, corrosion assessment, and predictive maintenance. The study synthesizes findings from a wide range of literature to identify key enabling technologies, including Internet of Things (IoT) sensors, data-driven modeling, computational fluid dynamics (CFD), and machine learning algorithms. Special attention is given to the integration of physics-based and data-driven models for improving the accuracy and reliability of digital twin frameworks. In addition, this paper proposes a unified reference architecture for pipeline digital twins, supported by a mathematical formulation of synchronization and a comparative synthesis of existing approaches. The review highlights how digital twins facilitate early fault detection and operational optimization by continuously synchronizing physical assets with their virtual counterparts. The review also emphasizes the importance of uncertainty-aware and reliability-informed digital twin frameworks for robust decision-making in safety-critical pipeline applications. Applications in subsea, oil and gas, and water distribution pipelines are explored, demonstrating the versatility of DT systems under different environmental and operational conditions. Despite significant progress, challenges remain in data integration, model validation, scalability, and cybersecurity. Furthermore, the lack of standardized architectures and interoperability frameworks limits widespread adoption. This paper concludes by outlining future research directions, including the development of hybrid modeling techniques, edge computing integration, and AI-driven autonomous decision systems. Overall, digital twin technology represents a paradigm shift in pipeline engineering, offering substantial potential to enhance safety, efficiency, and sustainability in complex infrastructure systems. Full article

The expansion of the Internet of Things (IoT) has increased the complexity of securing distributed systems against growing threats to data security, privacy, and reliability. Conventional centralised cybersecurity methods are often insufficient for environments characterised by scale, heterogeneity, and dynamic behaviour. This paper presents a structured review of Distributed Artificial Intelligence (DAI) for IoT security, focusing on how local, cooperative intelligence can support intrusion detection, anomaly recognition, secure data processing, and collaborative defence. We synthesise the current literature on Federated Learning (FL), Multi-Agent Systems, and related approaches, highlighting their benefits, limitations, and practical deployment constraints. Particular attention is given to critical infrastructure contexts, where resilience is essential for operational continuity and public safety. The review concludes by outlining key gaps and future research directions for DAI-enabled IoT security. Full article

The metalworking industry faces growing regulatory pressure to quantify and verify its product carbon footprint (PCF), driven by frameworks such as the European Union’s Carbon Border Adjustment Mechanism (CBAM) and emissions trading schemes (ETS). Traditional static accounting methods, based on generic emission factors and annual averages, are insufficient given the dynamic, multi-stakeholder nature of the sector’s supply chains. This study presents a systematic review conducted under the PRISMA 2020 protocol to identify, classify and critically evaluate the digital tools and technologies used to calculate, manage and verify PCF in this sector. From 495 records screened, 53 thematically relevant studies were analyzed and 5 sector-specific cases examined in depth. The results indicate that the Internet of Things (IoT) and smart sensor networks constitute the primary data-capture layer, while Machine Learning, Big Data and Digital Twins are the predominant processing technologies. Blockchain and verifiable digital credentials emerge as governance mechanisms that ensure the transparency, auditability and immutability of emissions inventories, enabling compliance through Digital Product Passports (DPPs). We conclude that digital decarbonization requires interoperable architectures integrating real-time capture, distributed traceability and common semantic standards; viability in small and medium-sized enterprises (SMEs) and interoperability across heterogeneous platforms remain the main research gaps. Full article

Background: The confluence of the Internet of Things (IoT) and the Physical Internet (PI) is a major accelerant of Logistics 4.0, which can provide significant upsides to supply chain performance. While prior reviews mainly concentrated on technological advancement and applications, limited consideration has been given to the mechanisms through which the PI-IoT system integration generates benefits, namely operational, economic, and environmental benefits. Methods: To fill this gap, this study undertakes a systematic literature review of 43 peer-reviewed studies guided by PRISMA. This study uses the Context, Intervention, Mechanism, Outcome (CIMO) framework to automatically conduct a mechanism-based synthesis of PI-IoT integration, unlike previous reviews. Results: The results show that enhanced operational and economic efficiency, as well as environmentally sustainable performance, can be achieved by using real-time visibility, predictive decision-making, collaborative resource optimization, intelligent automation, and adaptive system responsiveness. Conclusions: This review also proposes a conceptual framework, linking contextual conditions, technological interventions, mechanisms, and outcomes, while also identifying important research gaps and future research directions. This study offers essential findings that offer practical insights for researchers, logistics managers, and policymakers intending to implement collaborative, data-driven, and sustainable PI-IoT-enabled logistics systems. Full article

This systematic review summarises advances in artificial intelligence (AI) and machine learning (ML) for foodborne pathogen detection, covering applications in various technologies (AI-assisted microscopy, spectroscopy, biosensors and sensor-based systems), food supply chains, analytical performance, operational metrics and regulatory developments, addressing gaps in previous reviews limited to individual technologies or lacking regulatory analysis. Following PRISMA 2020 guidelines, Scopus, PubMed, and Web of Science were searched from 1 January 2010 to 25 June 2026 using a validated string. Inclusion criteria were explicit detection of a pathogen, clearly described AI/ML algorithm, study evaluation on food or supply chains, and quantitative validation metrics. Exclusion criteria were chemical-only studies, human-diagnostic studies, or purely theoretical studies. Given heterogeneity in the evidence, qualitative quality indicators were favoured over formal quantitative risk-of-bias tools, in distinction to internal cross-validation versus independent external validation. Key data were extracted using a standardised matrix, and after screening and snowballing, the final corpus consisted of 152 studies. CNN (Convolutional Neural Network)-based microscopy provides >99% accuracy in bacterial identification, SERS (Surface-Enhanced Raman Spectroscopy) and CNN 98.68% for pathogens and 99.85% for resistant strains. ML-driven biosensors show 80–100% prediction accuracy in the presence of environmental noise. Yet, performance drops dramatically on external validation, with models falling from 95% internal to 78–82% on independent test sets. Supply chain applications cover meat, dairy, seafood and produce, but most are still at pilot scale. The main constraints are data heterogeneity, lack of public benchmarks, matrix interference, non-standard validation protocols, and regulatory dissonance. However, the integration of AI with Internet of Things (IoT), blockchain and edge computing improves sensitivity, reduces false results and enables real-time monitoring despite the challenges. AI is a powerful decision-support tool that complements existing food safety controls rather than replacing them. To translate these technologies reliably into routine practice, effective implementation requires rigorous external validation and regulatory harmonisation. Full article

Thailand’s ceramic manufacturing tradition possesses a long and distinguished history, reflecting the nation’s rich cultural heritage, artistic excellence, and capacity for technological adaptation. Traditional Thai ceramics extend beyond their functional purposes, serving as important expressions of indigenous knowledge, craftsmanship, social values, and religious beliefs that have been transmitted across generations. While preserving their distinctive Thai characteristics, these ceramic traditions have continuously evolved through cultural exchanges with neighboring civilizations, particularly China and India, as well as later influences from the West. Among the various decorative techniques employed in Thai ceramics, stamping and hand-painted ornamentation are recognized as two of the most significant methods, contributing to the aesthetic and cultural value of ceramic works. These techniques enable ceramic products to embody both artistic expression and practical functionality by harmoniously integrating aesthetic design with reliable craftsmanship. In contemporary manufacturing environments, traditional stamping and painting methods are increasingly integrated with semi-automated processes and advanced ceramic machinery to enhance production efficiency while preserving cultural authenticity. This study proposes an Internet of Things (IoT)-based control system for ceramic stamping and painting machines, designed to support remote operation, real-time monitoring, and performance evaluation, with particular attention given to response time and error characteristics. By incorporating sensors, controllers, and networked communication technologies into ceramic manufacturing equipment, the proposed system establishes a meaningful connection between intelligent automation and traditional artistic practices. Full article

Technological and social development is desirable and even indispensable, which necessarily involves the restriction of new life situations within legal frameworks. European legislation has been visibly struggling with this problem in recent years, but the established/ongoing regulation may be an obstacle to development. Among other things, this includes the issue of regulating platform work. The emergence and spread of platform work has numerous advantages from an economic point of view, but from a legal point of view, the cautious regulation of this relatively new employment construction is not acceptable to the majority dealing with labour law. In our opinion, the relevant EU legislation is fundamentally flawed, as it basically seeks to answer the question of whether a given legal relationship is an employment relationship or not. The current binary classification might not be sufficient. Thus, the present study examines why platform work can be considered special and what are the labour law guarantees that are justified to be extended—at least as a rule—in this regard. To further investigate the practical risks of the current rules, a recent and relevant judgement of the Hungarian Supreme Court is also analysed in order to illustrate the uncertainties in litigation. The ruling demonstrates that traditional employment tests fail to recognise algorithmic control—including GPS surveillance, scheduling penalties, and unilateral remuneration determination—as indicators of subordination, while placing an insurmountable burden of proof on workers. This case empirically confirms the practical difficulties of the current binary classification. Our aim is to examine whether it is necessary to develop a minimum guarantee system that allows for easier transparency, greater legal certainty and a more uniform application of the law, unlike the current regulation. Full article

Control charts are used in statistical process control (SPC) to keep track of processes and identify changes in the way that they work. The control limits around the center line are uniform, which means they react the same way to changes going in either direction. In contrast, linear charts use imbalance to make it easier to identify individual data points. Therefore, using imbalance in the creation of control charts helps keep track of and maintain consistency with data that has a big impact on results when it goes beyond predetermined limits. In this study, we look at both one-sided and two-sided control charts by getting an explicit closed-form formula for the double-modified EWMA control chart’s average run length (ARL). The study is mostly about developing better ways to spot things using autoregressive fractionally integrated models and external variables (ARX and ARFIX) in the presence of exponential white noise. The ARL is used to test how well the proposed chart works in both modeling systems. The NIE method is used to prove that the explicit closed-form ARL formula works. The closed-form ARL expression is shown to be valid under the given ARX and ARFIX model assumptions, exponential white noise errors, stationarity conditions, and fixed one-sided or two-sided control limits. The results show that %RPC has a value below 10⁻⁶, and the computation times for the ARX and ARFIX models remain below 1.6 s and 3 s, respectively, after that point. To show how much better it is, the suggestion is compared to Type-EWMA control charts, such as classical and modified EWMA charts, in terms of run-length efficiency using ARL and SDRL, as well as overall efficiency by the relative index and with mean and standard deviation. The simulation study checks how well the proposed chart works in both symmetric two-sided and asymmetric one-sided frameworks. For the crude oil application, the one-sided upper control chart is used to detect abrupt upward price shifts, which may indicate precautionary demand shocks, market uncertainty, and risk spillovers to financial markets. According to the findings, the suggested chart is able to identify shifts at a faster rate than both traditional EWMA charts and modified EWMA charts, which demonstrates that it is beneficial in a real setting. Full article