Search Results (233)

Rail transport is central to achieving sustainable and energy-efficient mobility, and its digitalization is accelerating the adoption of condition-based maintenance (CBM) strategies. However, existing maintenance practices remain largely reactive or rely on limited rule-based diagnostics, which constrain safety, interoperability, and lifecycle optimization. This survey provides a comprehensive and structured review of Artificial Intelligence techniques applied to the preventive, predictive, and prescriptive maintenance of railway infrastructure. We analyze and compare machine learning and deep learning approaches—including neural networks, support vector machines, random forests, genetic algorithms, and end-to-end deep models—applied to parameters such as track geometry, vibration-based monitoring, and imaging-based inspection. The survey highlights the dominant data sources and feature engineering techniques, evaluates the model performance across subsystems, and identifies research gaps related to data quality, cross-network generalization, model robustness, and integration with real-time asset management platforms. We further discuss emerging research directions, including Digital Twins, edge AI, and Cyber–Physical predictive systems, which position AI as an enabler of autonomous infrastructure management. This survey defines the key challenges and opportunities to guide future research and standardization in intelligent railway maintenance ecosystems. Full article

This paper analyses the factors that affect CO₂ emissions in the BF-BOF steelmaking process using a dynamic econometric approach based on annual data from the Polish steel industry. The analysis commences with the estimation of a baseline dynamic model that describes the relationship between CO₂ emissions in the industry and investment allocations, crude steel production, and lagged CO₂ emissions. The baseline analysis illustrates the dominant feature of strong emission level persistence and poor tracking of selected conventional production-related factors. The analysis proceeds by extending the baseline results through additional consideration of technological factors, material composition factors, and resource use factors in the generation of CO₂ emissions. The additional factors include the use of coke, electricity consumption, fixed asset value, and the scrap ratio. The analysis indicates that these additional factors are essential in improving the accuracy of the modeling process and in clarifying the significance of material composition in CO₂ emissions in particular. The analysis further illustrates the critical result that increased use of electricity leads to high CO₂ emissions in the BF-BOF process. Further analysis indicates that increasing the use of steel scrap leads to substantial CO₂ reductions in the BF-BOF route and other steelmaking technologies. The results also show that CO₂ emissions in the BF-BOF process depend not only on production volume, but also on material composition and the technological structure of the process. In the context of the WFESF project, these findings provide evidence-based guidance for metal industry research by identifying priority levers for mitigation, particularly through improvements in process technology and scrap-based material substitution. Full article

(1) Background: In hospitals, mattresses are often relocated for cleaning or patient transfer, leading to mismatches between actual and recorded bed locations. Manual updates are time-consuming and error-prone, requiring an automatic localization system that is cost-effective and easy to deploy to ensure traceability and reduce nursing workload. (2) Purpose: This study presents a pragmatic, large-scale implementation and validation of a BLE-based localization system using RSSI measurements. The goal was to achieve reliable room-level identification of smart mattresses by leveraging existing hospital infrastructure. (3) Results: The system showed stable signals in the complex hospital environment, with a 12.04 dBm mean gap between primary and secondary rooms, accurately detecting mattress movements and restoring location confidence. Nurses reported easier operation, reduced manual checks, and improved accuracy, though occasional mismatches occurred when receivers were offline. (4) Conclusions: The RSSI-based system demonstrates a feasible and scalable model for real-world asset tracking. Future upgrades include receiver health monitoring, watchdog restarts, and enhanced user training to improve reliability and usability. (5) Method: RSSI–distance relationships were characterized under different partition conditions to determine parameters for room differentiation. To evaluate real-world scalability, a field validation involving 266 mattresses in 101 rooms over 42 h tested performance, along with relocation tests and nurse feedback. Full article

Material tracking and localization are key applications of Industry 4.0 in manufacturing process control. Traditional approaches—such as barcode or QR code identification and RTLS-based localization using RF/UWB, 5G or GPS–require a large and complex infrastructure. As an alternative, this paper proposes an IoT-based solution that combines short-range Bluetooth Low Energy (BLE) communication with LPWAN LoRaWAN networks. Hybrid solutions using LoRaWAN and BLE technologies already exist, but pure localization based on BLE tags can lead to ambiguous asset identification in geometrically dense scenarios. Our paper aims to solve this problem with an alternative concept called Associated LoRaWAN Sensors (ALSs). An ALS enables logical grouping and integration of heterogeneous LoRaWAN sensors, providing their own data or directly scanning BLE tags. Sensor data can be combined and supplemented with new information, data, and events, supported by application logic (use case). Although ALS represents a general concept that could be applicable to various use cases (such as warehouse monitoring, object tracking), our paper will focus mainly on material tracking and validation in manufacturing. For this purpose, we designed a specific ALS model that integrates a classic LoRaWAN BLE sensor with an additional LoRaWAN magnetic contact sensor. The magnetic contact switch can provide validation of exact position, in addition to localization by BLE tag. Experimental validation using BLE tags (Trax 10229) and LoRaWAN sensors (IoTracker3, Milesight WS301) demonstrates the usability of the ALS model in typical industrial scenarios. We also measured RSSI and evaluated the accuracy of tag localization (3 × 25 = 75 tests) for the worst-case scenario: material validation on a machine with a BLE tag distance of ~0.5 m. While the traditional approach showed up to a 20% failure rate, our ALS model avoided the issue of incorrect accuracy. An additional magnetic switch in ALS confirmed that the correct carrier with the associated tag is attached to the machine and eliminated incorrect localization. The results confirm that a hybrid model based on BLE and LoRaWAN scanning can reliably support material localization and validation without the need for dense RTLS infrastructures. Full article

The focus of this research is the automation of the synchronization process for track-recording vehicle signals in turnouts. Accurate synchronization of measurement signals is essential for assessing specific track sections—especially complex areas such as turnouts—and for enabling reliable time series for condition monitoring. Currently, the synchronization process is only partially automated, resulting in high levels of manual effort. With over 4000 turnouts on Austria’s main railways, full automation is important for ensuring efficiency and consistency of the synchronization process. Based on an analysis of 109 turnouts in the OeBB railways, the process begins with rough synchronization using mileage and curvature signals to eliminate invalid measurement runs. Subsequently, longitudinal level signals are synchronized within maintenance time blocks. These blocks include measurement runs with consistent signal characteristics between two maintenance interventions. The latest valid run then serves as reference for each block. Methods such as cumulative sum, Euclidean distance and cross-correlation are then employed to achieve fine synchronization. The results demonstrate the feasibility and efficiency of automated synchronization compared to manual methods, enabling more accurate condition assessment. This allows infrastructure managers to track turnout-specific quality indicators, integrate them into asset management systems, and develop predictive maintenance strategies. Full article

This paper proposes a novel methodology, based on machine learning and statistical models, for predicting hurricane-related losses to specific assets. Our approach uses three critical storm parameters typically tracked by meteorological agencies: maximum wind speed, minimum sea level pressure, and radius of maximum wind. The system categorizes potential damage events into three insurance-relevant classes: non-payable, partially payable, and fully payable. Three triggers for final payouts were designed: hybrid framework, standalone regression, and standalone non-linear regression. The hybrid framework combines two classification models and a non-linear regression model in an ensemble specifically designed to minimize the absolute differences between predicted and actual payouts (Total Absolute Error or TAE), addressing highly imbalanced and partially compensable events. Although this complex approach may not be suitable for all current contracts due to limited interpretability, it provides an approximate lower bound for the minimization of the absolute error. The standalone non-linear regression model is structurally simpler, yet it likewise offers limited transparency. This hybrid framework is not intended for direct deployment in parametric insurance contracts, but rather serves as a benchmarking and research tool to quantify the achievable reduction in basis risk under highly imbalanced conditions. The standalone linear regression provides an interpretable linear regression model optimized for feature selection and interaction terms, enabling direct deployment in parametric insurance contracts while maintaining transparency. These three approaches allow analysis of the trade-off between model complexity, predictive performance, and interpretability. The three approaches are compared using comprehensive hurricane simulation data from an industry-standard catastrophe model. The methodology is particularly valuable for parametric insurance applications, where rapid assessment and claims settlement are essential. Full article

Background/Objectives: Augmented Reality (AR) is promising in dental education, yet its impact on caries detection training remains underexplored. This study aimed to compare the effect of a web-based AR (WebAR) learning object with a content- and interface-matched interactive 2D presentation on undergraduate students’ motivation to learn caries detection. Methods: Two learning objects were expressly designed using a real patient’s dental records: a WebAR image-tracking experience (built with Zapworks Studio^®) and a 2D interactive presentation (built with Genially^®). The WebAR object showed the patient’s 3D dental arches with tooth-level hotspots linking clinical and radiographic media. The 2D comparator mirrored the same assets and navigation, restricting visualization to 2D. Third-year dental students were randomly assigned to either the AR or Genially^® (G) group. After completing ICDAS-based caries identification, participants completed the 12-item Reduced Instructional Materials Motivation Survey (RIMMS) and provided open-ended feedback. Group differences were tested with the Mann–Whitney U test (p < 0.05). Results: Eighty-five students completed the study (AR n = 46; G n = 39). The AR group achieved a higher total RIMMS score (4.14 vs. 3.53 on a 5-point scale; p < 0.001), with significantly higher means in Attention, Confidence, Satisfaction, and Relevance (p < 0.05). Open-ended comments were more positive with AR (75.8% vs. 31.0%), while graphics-related complaints were more frequent with the Genially^® resource (34.5% vs. 75.0%). Conclusions: WebAR achieved higher RIMMS motivation scores than a content-matched interactive presentation. Adding 3D spatial interaction to otherwise equivalent materials can enhance learners’ motivation for caries detection training, while remaining low-cost and scalable. Full article

Taxation on cryptocurrency is becoming critical in global fiscal governance as digital assets adapt to the modern reality of existing outside of traditional regulatory constructs. Theoretical and practical understanding of cryptocurrency taxation is quite new, and so a systematic review was designed to present the most recent empirical research evidence on the legal, fiscal and behavioral aspects of cryptocurrency taxation from across the globe. Using the PRISMA-2020 guidelines, a structured search was applied to the Scopus database on 21 May 2025, with the search terms “crypto-currency”, “cryptoasset” and “taxation.” The inclusion criteria consisted of original research articles published between the years of 2020 and 2025 in English or Spanish, that could be accessed via institutional library support, and that were related to taxation, legal regulation and/or compliance. Out of the original identified 224 records, 36 met the eligibility criteria after screening and verification through seven different stages of review. Socially, five themes were produced by the findings: legal ambiguity surrounding fiscal treatment, limited tax literacy and compliance issues, macroeconomic and monetary issues, application of digital technologies for fiscal tracking, and environmental repercussions from crypto mining. Many countries do not have any coherent tax frameworks to govern the risk that emerges from cryptocurrency taxation, creating uncertainty for both regulators and investors. The findings outlined in this systematic review point to the urgent need for creating a coherent approach to cryptocurrency taxation based on definitions, digital approaches to traceability, and tax literacy compliance strategies. In order to create effective cryptocurrency taxation, there must be a base balance between ensuring innovation, fiscal responsibility, transparency, equity and sustainability in the developing digital economy. Full article

The tracking of assets or cargo is one of the main objectives of global logistics and transportation systems, ensuring operational efficiency, security, and timeliness. Currently, battery-operated GPS (Global Positioning System)-based tracking devices are used for this purpose. The main shortcoming of these devices is the lifetime of the batteries because they cannot be replaced or recharged, or because this is simply not economically feasible. Therefore, efficient methods are needed to prolong battery life as much as possible. Various existing energy-saving techniques can be applied to solve this problem. However, none of these consider situations in which multiple tracking devices are transported together and can cooperate to further increase their energy efficiency. In this study, we propose and evaluate the novel lightweight peer-to-peer energy-saving method for nearby wireless battery-powered trackers based on their cooperation. The proposed method is based on the short-range BLE (Bluetooth Low Energy) device discovery mechanism and the dynamic election of the leader tracker (with the highest battery capacity) to report the location of its own and other neighboring trackers to the central server. The experimental evaluation of the proposed method shows that, compared to the traditional approach, where each tracker sends its location individually, the proposed method allows a reduction in the average battery charge required for one position report from 19% to 240% per each cooperating tracker. The average energy consumption for one location report per node decreased from 4.68 mWh using the traditional approach to 3.93 mWh for 2 cooperating devices and 1.92 mWh for 15 cooperating devices. Full article

This article presented a method for grading and visualising corrosion in steel pedestrian bridges using Building Information Modelling (BIM). Traditional inspection methods are often manual and subjective, which reduces their reliability and repeatability. To enhance the recording and reporting of inspection results, a five-level corrosion severity grading system was developed using matched photographic data from two inspection campaigns conducted in February 2024 and April 2025. The grades were assigned based on visual signs, including surface rust, coating damage, and flaking. A Dynamo script was used to link each grade to the corresponding elements in a Revit model using colour overrides. The proposed approach enables corrosion data to be integrated into the BIM environment in a clear, structured manner. This helps engineers assess the structure’s condition, monitor changes over time, and make informed maintenance decisions. The workflow was demonstrated using case studies from a steel pedestrian bridge in Aveiro, Portugal. The method is adaptable for future digital twin applications and supports the development of BIM-based tools for bridge asset management. The workflow was applied to over 2600 elements, with 75 visually degraded cases identified and classified into five grades, demonstrating the method’s feasibility for systematic corrosion tracking. The proposed workflow was tested on a coastal steel bridge and could be generalised to other bridges with similar environmental conditions. Full article

To address the triple challenges of data sparsity, highly nonlinear dynamics, and maneuver uncertainty in tracking non-cooperative targets in cislunar space, we propose a collaborative framework combining Particle Filter (PF) and Unscented Kalman Filter (UKF). This framework optimizes search efficiency through a two-phase strategy: in the search phase, PF constructs the target reachable domain and leverages undetected information to dynamically shrink the search scope; upon target detection, the framework switches to UKF for high-precision and low-overhead tracking. To overcome the computational bottleneck in high-dimensional reachable domain integration, we integrate a non-product-type Designed Quadrature (DQ) method—one that generates minimal quadrature point sets to replace traditional Monte Carlo sampling by matching the moment conditions of mixed distributions via Gauss–Newton optimization. Distinct from existing single-filter or reachability modeling approaches, the key novelties of this work lie in a two-phase PF-UKF switching framework tailored to the unique cislunar environment resolving the trade-off between search capability and computational efficiency and integration of the non-product DQ method to break the dimensionality curse in high-dimensional reachable domain computation ensuring both moment-matching accuracy and real-time performance. This work holds potential to support space domain awareness and cislunar mission safety: reliable tracking of non-cooperative targets is a key prerequisite for avoiding collisions, safeguarding space assets, and enabling effective space defense, and the proposed framework provides a feasible technical path for this goal through simulation validation. Simulations demonstrate that on a three-dimensional Distant Retrograde Orbit (DRO) observation platform, successful recapture of cislunar transfer orbit targets can be achieved. Under fifth-order accuracy conditions, the system exhibits a position error of $3.745\times {10}^{-1}\mathrm{km}$ and a velocity tracking error of $9.703\times {10}^{-3}\mathrm{m}/\mathrm{s}$ for target search-and-tracking tasks, with a system response time of 1.8343 h. Compared with the traditional PF + numerical integration method, our proposed PF-UKF framework achieves an 86.7% reduction in time cost and a 24.1% reduction in position error. Full article

This paper presents a real-time, output-only structural health monitoring framework that integrates damage-sensitive cepstral features with a streaming Long Short-Term Memory (LSTM) network for automated damage detection. Acceleration time histories are segmented into overlapping windows, converted into cepstral coefficients, and processed sequentially by a stacked LSTM architecture with state carry-over. This design preserves temporal dependencies while enabling low-latency inference suitable for continuous monitoring. The framework was evaluated under a strict zero-shot setting on the full-scale Z24 Bridge benchmark, in which no training or calibration data from the bridge were used. Our results show that the proposed approach can reliably discriminate staged damage states and track their progression using only vibration measurements. By combining a well-established spectral feature representation with a streaming sequence model, the study demonstrates a practical pathway toward deployable, data-driven monitoring systems capable of operating without retraining on each individual asset. Full article

Background: Effective asset tracking and monitoring are critical for modern warehouse management. Methods: In this paper, we present a real-time warehouse monitoring system that leverages ceiling-mounted cameras, computer vision-based object detection, a knowledge-graph based world model. The system is implemented in two architectural configurations: a distributed setup with edge processing and a centralized setup. Results: Experimental results demonstrate the system’s capability to accurately detect and continuously track common warehouse assets such as pallets, boxes, and forklifts. This work provides a detailed methodology, covering aspects from camera placement and neural network training to world model integration and real-world deployment. Conclusions: Our experiments show that the system achieves high detection accuracy and reliable real-time tracking across multiple viewpoints, and it is easily scalable to large-scale logistics and inventory applications. Full article

Understanding landscape fragmentation is crucial to explore comprehensive land-use–land-cover (LULC) change within fast-growing urbanisation. While LULC change is a global concern, limited research explores landscape fragmentation along river and road infrastructure in high-density riverine contexts. This study addresses this gap through understanding dynamic landscape fragmentation patterns to track LULC in the Ciliwung River, Indonesia, from 1990 to 2020. The research employed remote sensing, GIS, R programming with Landsat data, Normalised Difference Vegetation Index (NDVI) values, buffering, and landscape metrics. The findings show minimal fragmentation was concentrated downstream near Jakarta, while significant fragmentation, manifesting as green loss, occurred in the midstream. Buffer analysis showed high green loss in the upstream segment both near the river and roads, particularly within a 0–400 m buffer. However, landscape metrics identified changes in the midstream close to the river buffer (0–200 m) indicating that riparian green spaces in this area persist as relatively large but ecologically unconnected “chunks”. The stability of these remaining patches makes them a crucial asset for targeted restoration. These findings contribute to a deeper understanding of how river and road networks influence the change, highlighting the integral role of remote sensing and GIS in monitoring LULC change for natural preservation. Full article

Power transformers are critical assets in modern power grids, where failures can lead to significant operational disruptions and financial losses. Dissolved Gas Analysis (DGA) is a key sensor-based technique widely used for condition monitoring, but traditional diagnostic approaches rely on deterministic thresholds that overlook uncertainty in degradation dynamics. This paper proposes a probabilistic framework for Prognostics and Health Management (PHM) of power transformers, integrating self-adaptive Auto Regressive Integrated Moving Average modeling with a probabilistic reformulation of Duval’s graphical methods. The framework enables automated estimation of fault types and failure likelihood directly from DGA sensor data, without requiring labeled datasets or expert-defined rules. Dissolved gas dynamics are forecasted using time-series models with residual-based uncertainty quantification, allowing probabilistic fault inference from predicted gas trends without assuming deterministic persistence of a specific fault type. A sequential pipeline is developed for real-time fault tracking and reliability assessment, aligned with IEC, IEEE, and CIGRE standards. Two case studies validate the method: one involving gas loss in an experimental setup and another examining thermal degradation in a 345 kV transformer. Results show that the framework improves diagnostic reliability, supports early fault detection, and enhances predictive maintenance strategies. By combining probabilistic modeling, time-series forecasting, and sensor-based diagnostic inference, this work contributes a practical and interpretable PHM solution for sensor-enabled monitoring environments in modern power grids. Full article