Search Results (1,942)

Accurately forecasting demand for intermittent items is essential for effective inventory control, improved service levels, and cost reduction. This study focuses on highly sparse, irregular, and volatile demand patterns and proposes a generalizable multi-source data-driven framework for intermittent demand forecasting, using automotive spare parts as a representative application scenario. The proposed framework integrates Transformer networks, multi-graph convolutional networks (GCNs), and a Mamba-based feature fusion module. The Transformer captures long-term temporal dependencies in historical demand sequences, while the multi-graph GCN incorporates prior knowledge—including traffic geography, socioeconomic indicators, and environmental attributes—to model spatial correlations across multiple supply nodes. The Mamba-based fusion module then integrates temporal and spatial features into a unified representation, enhancing predictive accuracy and robustness. Extensive experiments on real-world datasets of automotive spare parts in China show that the proposed framework exhibits competitive and often superior performance compared with TiDE, FSNet, Informer, and DLinear across multiple forecasting horizons (3-, 6-, and 9-step), as measured by RMSE, MAE, and $R^2$. The proposed approach provides a practical and adaptable solution for forecasting intermittent demand, offering valuable support for dynamic inventory management. Full article

Transportation hubs serve as critical convergence points for traffic, information, and energy flows. However, their energy systems are characterized by high consumption randomness, significant power flow fluctuations, and geographically dispersed source and load nodes. These features pose challenges for integrating distributed renewable energy and often lead to high energy cost issues. Additionally, accommodating distributed photovoltaic (PV) is further complicated by grid corridors and high investment expenditure. To address these issues, this paper proposes a two-stage optimization model for a multi-node interconnected architecture for transportation hubs. In the first stage, a greedy algorithm determines a fixed connection topology, considering distance constraints and connection port limits to ensure engineering feasibility. The second employs linear programming to optimize real-time power allocation. This approach precomputes connection relationships, significantly reducing evaluation time and enabling efficient processing of operational data from multiple nodes. A case study confirms that the proposed method can increase PV consumption by 38.71%, with optimization evaluated on a millisecond scale. By inputting node generation, load, and distance data in prescribed format, the model outputs actionable planning results for flexible interconnection projects. Full article

The present work explores the effects of dental splints from a mechanical standpoint, aiming to provide a practical tool for the surgical decision-making process regarding splint cross-section dimensions. Our investigation centers on the anatomical structure of a pentamorphic dental arch encompassing central and lateral incisors and one canine on each side. Using parametric in silico models built up by means of an ad-hoc procedure, geometry, material properties, and boundary conditions are defined on a parametric anatomical model that can be tailored using RX-derived geometrical information. Two general cases have been considered, one with the splint and the other splintless, and a sensitivity analysis has been performed by varying the splint section height and thickness. The results show the diminishing mobility at the apex and basis of the diseased incisors, demonstrating the effectiveness of the periodontal treatment. Moreover, the stress due to physiological loads moves away from diseased teeth toward the healthy ones due to the splint effects, focusing on the splint–glue–canine contact zone and highlighting the crucial role played by the canine in fixing the entire dental structure. To establish a preliminary mechanical assessment of the dental structure’s safety and to confine its actual value within a mechanically reasonable range, a synthetic “traffic-light” indicator of stress-based failure risk is proposed. It is felt that the tool proposed in this study can help surgeons assess the pre-operative patient-specific mechanical effects of the splint treatment, driving the design and choice of periodontal splints. By linking splint geometry to mechanical safety via a stress-based indicator, the method supports the optimized design and selection of splints, improving treatment reliability while preserving comfort and clinical effectiveness. Full article

Particulate Matter (PM) is a type of air pollution that poses risks to human health, the environment, and property. Among the various PM types, PM₁₀ is particularly significant, as it acts as a vector for numerous hazardous trace elements that can negatively impact human health and the ecosystem. Identifying potential sources of PM₁₀ and quantifying their impact on ambient concentrations is crucial for developing efficient control strategies to meet threshold values. Receptor modeling, which identifies sources using chemical species information derived from PM samples, has been widely used for source apportionment. In this study, PM₁₀ samples were collected over three periods (April, May, and June 2021), each lasting 16 days, using semi-automatic dust sampling systems at two sites in Biga, Canakkale, Turkiye. The relative contributions of different source types were quantified using EPA PMF (Positive Matrix Factorization) based on 35 elements comprising PM₁₀. As a result of the analysis, five source types were identified: crustal elements/limestone/calcite quarry (64.9%), coal-fired power plants (11.2%), metal industry (9%), sea salt and ship emissions (8.5%), and road traffic emissions and road dust (6.3%). The distribution of source contributions aligned with the locations of identified sources in the region. Full article

With the rapid advancement of autonomous driving technology, Traffic Sign Detection and Recognition (TSDR) has become a critical component for ensuring vehicle safety. However, existing TSDR systems still face significant challenges in accurately detecting partially occluded traffic signs, which poses a substantial risk in real-world applications. To address this issue, this study proposes a comprehensive solution from three perspectives: data augmentation, model architecture enhancement, and dataset construction. We propose an innovative network framework tailored for occluded traffic sign detection. The framework enhances feature representation through a dual-path convolutional mechanism (DualConv) that preserves information flow even when parts of the sign are blocked, and employs a spatial attention module (SEAM) that helps the model focus on visible sign regions while ignoring occluded areas. Finally, we construct the Jinzhou Traffic Sign (JZTS) occlusion dataset to provide targeted training and evaluation samples. Extensive experiments on the public Tsinghua-Tencent 100K (TT-100K) dataset and our JZTS dataset demonstrate the superior performance and strong generalisation capability of our model under occlusion conditions. This work not only advances the robustness of TSDR systems for autonomous driving but also provides a valuable benchmark for future research. Full article

Distributed photovoltaic (DPV) systems present a cost-effective and sustainable industrial energy solution, yet their reliable monitoring faces significant technological constraints. This paper proposes a hierarchical optimization framework that integrates hysteresis-based traffic shaping at the network layer with Large Language Model (LLM)-driven diagnostics at the application layer. The proposed dynamic algorithm minimizes latency and downtime by prioritizing critical fault data. Priority-based scheduling ensures this critical data is transmitted preferentially over routine sensor readings. At the application layer, the system utilizes physics-informed prompt engineering to perform zero-shot root cause analysis, circumventing the training data requirements of traditional classifiers. Under a 10 Mbps gateway bandwidth, our method achieves a 46.08% to 49.87% reduction in P50 latency compared to traditional approaches. Moreover, the LLM-powered diagnostic system provides detailed assessments, enabling precise fault diagnosis for DPV systems. Full article

This paper presents a novel deep learning framework for traffic crash prediction that leverages graph-based representations to model complex interactions among road users. At its core is a dynamic Graph Attention Network (GAT), which abstracts road users and their interactions as evolving nodes and edges in a spatiotemporal graph. Each node represents an individual road user, characterized by its state as features, such as location and velocity. A node-wise Long Short-Term Memory (LSTM) network is employed to capture the temporal evolution of these features. Edges are dynamically constructed based on spatial and temporal proximity, existing only when distance and time thresholds are met for modeling interaction relevance. The GAT learns attention-weighted representations of these dynamic interactions, which are subsequently used by a classifier to predict the risk of a crash. Experimental results demonstrate that the proposed GAT-based method achieves 86.1% prediction accuracy, highlighting its effectiveness for proactive collision risk assessment and its potential to inform real-time warning systems and preventive safety interventions. Full article

Improving cyclist safety conditions in the urban context is a key strategy to promote sustainable transport modes and reduce noise and environmental pollution. Recent plans have also addressed this point. In September 2020, the UN General Assembly declared the Decade of Action for Road Safety 2021–2030, aiming to reduce the number of road deaths by at least half. To achieve this task and highlight the risk factor, after collecting and pre-processing cyclist crash data in the city of Rome between 2013 and 2020, Random Forest and Ordered Logistic Regression models are proposed. The crash dataset is also enriched with vehicular speed and flows, and geographical information. A DBSCAN Clustering Analysis is also proposed to identify anomalous areas in the city. The findings show that the presence of cycle paths, the presence of anthropic activities, such as shops, schools, and universities, play a risk mitigation role. Conversely, vehicular speed and heavy vehicles emerge as the main detected risk factors. Finally, spatial analysis indicates that commercial activities reduce cycle path safety due to complex interactions with other road users. Furthermore, historic areas present unique risks driven by pedestrian flows and poor road surfaces, despite low vehicular traffic. Full article

The rapid growth of Unmanned Aircraft Systems (UASs) and Advanced Air Mobility (AAM) presents significant integration and safety challenges for the National Airspace System (NAS), often relying on disconnected Air Traffic Management (ATM) and Unmanned Aircraft System Traffic Management (UTM) practices that contribute to airspace incidents. This study evaluates Geographic Information Systems (GISs) as a unified, data-driven framework to enhance shared airspace safety and efficiency. A comprehensive, multi-phase methodology was developed using GIS (specifically Esri ArcGIS Pro) to integrate heterogeneous aviation data, including FAA aeronautical data, Automatic Dependent Surveillance–Broadcast (ADS-B) for crewed aircraft, and UAS Flight Records, necessitating detailed spatial–temporal data preprocessing for harmonization. The effectiveness of this GIS-based approach was demonstrated through a case study analyzing a critical interaction between a University UAS (Da-Jiang Innovations (DJI) M300) and a crewed Piper PA-28-181 near Purdue University Airport (KLAF). The resulting two-dimensional (2D) and three-dimensional (3D) models successfully enabled the visualization, quantitative measurement, and analysis of aircraft trajectories, confirming a minimum separation of approximately 459 feet laterally and 339 feet vertically. The findings confirm that a GIS offers a centralized, scalable platform for collating, analyzing, modeling, and visualizing air traffic operations, directly addressing ATM/UTM integration deficiencies. This GIS framework, especially when combined with advancements in sensor technologies and Artificial Intelligence (AI) for anomaly detection, is critical for modernizing NAS oversight, improving situational awareness, and establishing a foundation for real-time risk prediction and dynamic airspace management. Full article

Traffic state estimation (TSE) is a core task in intelligent transportation systems (ITSs) that seeks to infer key operational parameters—such as speed, flow, and density—from limited observational data. Existing methods often face challenges in practical deployment, including limited estimation accuracy, insufficient physical consistency, and weak generalization capability. To address these issues, this paper proposes a hybrid estimation framework that integrates multi-agent reinforcement learning (MARL) with the Lighthill–Whitham–Richards (LWR) traffic flow model. In this framework, each roadside detector is modeled as an agent that adaptively learns fundamental diagram (FD) parameters—the free-flow speed and jam density—by fusing local detector measurements with global CAV trajectory sequences via an interactive attention mechanism. The learned parameters are then passed to an LWR solver to perform sequential (rolling) prediction of traffic states across the entire road segment. We design a reward function that jointly penalizes estimation error and violations of physical constraints, enabling the agents to learn accurate and physically consistent dynamic traffic state estimates through interaction with the physics-based LWR environment. Experiments on simulated and real-world datasets demonstrate that the proposed method outperforms existing models in estimation accuracy, real-time performance, and cross-scenario generalization. It faithfully reproduces dynamic traffic phenomena, such as shockwaves and queue waves, demonstrating robustness and practical potential for deployment in complex traffic environments. Full article

(1) Background: Urban traffic micro-environments show strong spatial and temporal variability. Short and intensive campaigns remain a practical approach for understanding exposure patterns in complex environments, but they need clear and interpretable summaries that are not limited to simple site or time segmentation. (2) Methods: We carried out a multi-site campaign across five traffic-affected micro-environments, where measurements covered several pollutants, gases, and meteorological variables. A machine learning framework was introduced to learn interpretable operational regimes as recurring multivariate states using clustering with stability checks, and then we evaluated their added explanatory value and cross-site transfer using a strict site hold-out design to avoid information leakage. (3) Results: Five regimes were identified, representing combinations of emission intensity and ventilation strength. Incorporating regime information increased the explanatory power of simple NO₂ models and allowed the imputation of missing H₂S day using regime-aware random forest with an $R^2$ near 0.97. Regime labels remained identifiable using reduced sensor sets, while cross-site forecasting transferred well for NO₂ but was limited for PM, indicating stronger local effects for particles. (4) Conclusions: Operational-regime learning can transform short multivariate campaigns into practical and interpretable summaries of urban air pollution, while supporting data recovery and cautious model transfer. Full article

Conventional sensing expends energy at three stages: powering dedicated sensors, transmitting measurements, and executing computationally intensive inference. Wireless sensing re-purposes WiFi channel state information (CSI) inherent in every packet, eliminating extra sensors and uplink traffic, though reliance on deep neural networks (DNNs) often trained and run on graphics processing units (GPUs) can negate these gains. This review highlights two core energy efficiency levers in CSI-based wireless sensing. First ambient CSI harvesting cuts power use by an order of magnitude compared to radar and active Internet of Things (IoT) sensors. Second, integrated sensing and communication (ISAC) embeds sensing functionality into existing WiFi links, thereby reducing device count, battery waste, and carbon impact. We review conventional handcrafted and accuracy-first methods to set the stage for surveying green learning strategies and lightweight learning techniques, including compact hybrid neural architectures, pruning, knowledge distillation, quantisation, and semi-supervised training that preserve accuracy while reducing model size and memory footprint. We also discuss hardware co-design from low-power microcontrollers to edge application-specific integrated circuits (ASICs) and WiFi firmware extensions that align computation with platform constraints. Finally, we identify open challenges in domain-robust compression, multi-antenna calibration, energy-proportionate model scaling, and standardised joules per inference metrics. Our aim is a practical battery-friendly wireless sensing stack ready for smart home and 6G era deployments. Full article

Sustainable urban mobility necessitates traffic regimes that enhance operational efficiency and improve traffic safety and flow stability; the rise in intelligent connected vehicles (ICVs) provides a salient mechanism to meet this imperative. This paper aims to investigate the mixed traffic flow characteristics in an intelligent connected environment, using one-way single-lane, double-lane, and three-lane straight highways as modeling objects. Combining the different driving characteristics of human-driven vehicles (HDVs) and ICVs, a single-lane mixed traffic flow model and a multi-lane mixed traffic flow model are established based on the intelligent driver model (IDM) and flexible symmetric two-lane cellular automata model (FSTCAM). The mixed traffic flow in the intelligent connected environment is then simulated using MATLAB R2021a. The research results indicate that the integration of ICVs can improve the speed, flow, and critical density of traffic flow. The increase in the proportion of ICVs can reduce the congestion ratio and speed difference between front and rear vehicles at the same density. As the proportion of ICVs increases, the frequency of lane-changing for HDVs gradually increases, while the frequency of lane-changing for ICVs gradually decreases. The overall lane-changing frequency shows a trend of first increasing and then decreasing. In addition, with the continuous infiltration of ICVs, the area of road congestion gradually decreases, and congestion is significantly alleviated. The speed fluctuation of following vehicles gradually decreases. When the infiltration rate reaches a high level, vehicles travel at a stable speed and remain in a relatively steady state. The findings substantiate the potential of ICV-enabled operations to advance efficiency-oriented and stability-enhancing urban mobility and to inform evidence-based traffic management and policy design. Full article

Although pre-service teachers play a crucial role in promoting safe mobility among children, there are no validated instruments to assess their civic competencies, knowledge, and behaviors in road safety education. Existing questionnaires primarily target the general population or college students, and thus it remains unclear whether future teachers are adequately prepared to deliver road safety education. This study aims to design and validate a tool to assess pre-service teachers’ behavior in situations related to traffic safety and their knowledge of road safety education.. The designed tool is a questionnaire made up of 32 items distributed across five dimensions. The questionnaire’s content was validated through the judgment of eight experts, who ensured the relevance and adequacy of its items. The confirmatory factor analysis of data obtained from a pilot sample was used to examine the questionnaire’s structure, and reliability was assessed using Cronbach’s alpha coefficient. After analyzing the responses of 388 participants, the results suggest that the questionnaire’s overall structure is adequate and satisfactory reliability coefficients were obtained. The confirmatory factor analysis supported the proposed four-factor structure, indicating good model fit. These findings suggest that a valid and reliable diagnostic tool can identify the road safety training needs of future teachers and inform curriculum design and targeted educational interventions to enhance road safety competencies in schools. Full article

This study aimed to explore the factors influencing right-turning vehicles’ yielding behavior for crossing e-bikes at signalized intersections to improve safety for crossing e-bikes. Videos of different intersections were obtained through manual video recording and drone aerial photography. Spatiotemporal information data for right-turning vehicles and straight-through e-bikes were extracted through Tracker 6.0 software. Right-turning vehicle yielding decisions were categorized into three types: no yielding, decelerating to yield, and stopping to yield. Five potential variables influencing yielding decisions were selected: personal attributes of e-bike riders, traffic characteristics of e-bikes, traffic characteristics of right-turning vehicles, road characteristics, and right-turning vehicle–e-bike interaction influence characteristics. A multiple ordered logistic regression model was established to predict right-turn vehicle yielding decisions. Simultaneously calculating the OR (Odds Ratio) value reveals the likelihood of increased yielding probability under varying factors. For every one-unit increase in the number of crossing e-bikes, the yielding probability increases to 1.002 times the original value; for every one-unit increase in the average speed of right-turning vehicles, the yielding probability decreases to 0.406 times the original value; for every one-unit increase in the average crossing speed of e-bikes, the yielding probability increases to 1.737 times the original value. Compared with the straight + right-turn lane, a dedicated right-turning lane increases the yielding probability of right-turning vehicles to 4.2 times, and compared with not occupying a crosswalk, illegally occupying a crosswalk decreases the yielding probability of right-turning vehicles to 0.356 times. These findings offer valuable insights for enhancing the safety of e-bikes crossing signal-controlled intersections. Full article