Search Results (2,430)

The vehicle routing problem with time windows (VRPTW) is a core challenge in logistics optimization, requiring the minimization of transportation costs under constraints such as time windows and vehicle capacity. Deep reinforcement learning (DRL) provides an effective approach for solving such complex combinatorial optimization problems. However, existing DRL methods still suffer from shortcomings, including insufficient modeling of spatiotemporal correlations among customer nodes, inadequate capture of path temporal dependencies, and policy exploration prone to local optima. To address these issues, this paper proposes an end-to-end hybrid DRL framework: the encoder employs a graph attention network (GATv2) with adaptive gating to effectively model the coupling between customer spatial proximity and time window constraints; the decoder integrates multi-head attention (MHA) and a dynamic context-aware long short-term memory network (LSTM) to synergistically enhance the overall quality and constraint feasibility of route solutions; during the training phase, an improved proximal policy optimization (PPO) algorithm and a constraint-aware composite reward function are used to enhance optimization stability. Experiments on random instances, Solomon benchmark datasets, and real-world logistics datasets show that, compared to mainstream DRL methods and classical heuristic algorithms, the proposed framework reduces transportation costs by 2–10%, achieves a demand fulfillment rate exceeding 99%, and exhibits a performance degradation of only 3.2% in cross-distribution testing. This study provides an integrated DRL solution paradigm for combinatorial optimization problems with complex constraints, promoting the application of DRL in the field of intelligent logistics. Full article

This paper addresses the issue of coordination failures in fully mechanized mining equipment under complex operating conditions, which can lead to operational abnormalities and safety hazards. We systematically analyze the dynamic coordination relationships within the equipment system across three dimensions: temporal, spatial, and geometric. Centered on the coal flow, we establish a comprehensive “mining–transportation” coordination mathematical model covering the entire production process from the coal flow cut off by the shearer to the coal flow transported out by the conveyor. Building upon this foundation, a deep learning prediction method integrating long short-term memory (LSTM) and transformer architectures is proposed to construct an intelligent prediction model for the shearer traction speed. This model effectively captures temporal features and long-term dependencies within equipment operation data, enabling the prediction of critical operational parameters for fully mechanized mining systems. It significantly enhances the early identification and warning capabilities for equipment coordination failure states. The experimental results based on the operational data of fully mechanized mining systems show that the LSTM–Transformer model performs excellently in the prediction of traction speed. The mean square error (MSE) of prediction reached 0.041, the mean absolute error (MAE) was 0.122, and the coefficient of determination (R²) was 0.996, fully demonstrating the advantages of the model in terms of prediction accuracy and stability. This article provides a theoretical basis and technical support for the judgment of the operating status of coal mine working faces and the early warning of accident risks, which is of great significance for promoting the intelligent construction of coal mines. Full article

The accurate detection of electricity theft is crucial for reducing non-technical losses in smart grids. However, many existing data-driven methods rely on a single data modality, such as either the raw 1D consumption sequence or its transformed 2D image. This single-modality approach may not fully capture the complex spatio-temporal patterns associated with fraudulent behavior. To address this limitation, this paper proposes a novel detection method that integrates Markov Transition Fields (MTFs) with a hybrid neural network. First, this approach uses MTF to convert 1D time-series consumption data into 2D feature images, which enhances state-transition patterns. A parallel Residual Network and Long Short-Term Memory (ResNet-LSTM) architecture is then designed to simultaneously extract global temporal features from the original 1D data and local spatial features from the MTF images, with their fused representation used for classification. Experimental validation on a real-world dataset from the State Grid Corporation of China (SGCC)—comprising 6000 users over 304 days—demonstrates the effectiveness of our approach. The proposed model achieves a detection accuracy of 94.0% on an independent test set of 1200 users, significantly outperforming several state-of-the-art single-modality benchmarks. This work provides a new technical method for intelligent electricity theft prevention system. Full article

Background/Objectives: Early intervention in the prodromal stages of dementia is a primary focus of contemporary research, as delaying clinical progression may have a substantial public health impact. Citicoline, an endogenous precursor of phosphatidylcholine and acetylcholine, has been proposed as a nutritional compound with potential relevance to multiple cognitive domains. However, real-world evidence regarding its specific contributions in prodromal dementia populations is limited. This study was conducted to examine cognitive, functional, and emotional outcomes associated with the use of an oral citicoline combined preparation in individuals with prodromal dementia and early Alzheimer’s type cognitive decline. Methods: This was a two-centre, retrospective, observational, real-world cohort study. A cohort of 100 patients receiving a combined oral citicoline preparation and 50 age-matched healthy controls were evaluated at baseline and followed for 6–9 months. Participants underwent comprehensive neuropsychological assessments that evaluated domains including executive function, attention, processing speed, working memory, visual-spatial and verbal memory, fluency, general cognition, and mood. Standardized instruments included Stroop indices, Trail Making Tests A/B, SDMT, SPART-based measures, SBST, fluency tasks, the Boston Naming Test, and MoCA. Statistical analyses included age-adjusted and education-level-stratified comparisons. Results: Use of the citicoline combined preparation was associated with improvements in several cognitive domains, including executive functions, processing speed, working memory, visual-spatial memory, and both semantic and episodic fluency (all p < 0.05). Functional memory scanning and global cognition also showed improvement over the observation period. Significant differences between groups were observed at baseline and follow-up for multiple cognitive indices (most p < 0.001). Mood outcomes were more favorable in the citicoline combined preparation group, with reductions in depressive and anxiety symptoms. Age-adjusted models identified age as an important covariate, and participants with lower educational levels demonstrated comparatively greater cognitive gains. Conclusions: In this real-world observational study, use of an oral citicoline combined preparation was associated with multidomain improvements in cognitive and mood-related outcomes in individuals with prodromal dementia/early Alzheimer-type decline. Given the observational design, these findings should be considered exploratory and require confirmation in prospective randomised controlled trials. Full article

Multi object tracking (MOT) in UAV imagery is challenged by weak feature representation of small objects due to limited resolution, which leads to frequent missed detections. However, enhancing small object features often amplifies background noise and increases false positives. To address this contradiction, we propose the Temporal Aware Small Object Enhancement Network (TASONet), which integrates spatial enhancement and temporal modeling for robust tracking. The Small Object Enhancement (SOE) module combines depthwise separable convolutions with contrast-aware attention mechanisms (SimAM and LCDAttn) to improve local discriminability. It further incorporates the Small Target Enhancement Path (STEP), which uses motion-difference cues and a confidence adaptive suppression strategy to strengthen spatial features while mitigating noise. The Temporal Enhancement Module (TEM), consisting of Temporal Feature Alignment (TFA) and a Target Memory Unit (TMU), aggregates multi-frame information through adaptive inter-frame fusion and memory of high confidence historical features, improving temporal consistency and reducing false positives potentially introduced by SOE. Experiments show that TASONet achieves significant gains over state-of-the-art methods: on UAVDT, MOTA increases from 68.33 to 75.97 and IDF1 from 83.50 to 88.51; on VisDrone-MOT, MOTA rises from 61.15 to 73.52 with an IDF1 of 88.83. These results validate the effectiveness of jointly enhancing spatial features and temporal coherence for UAV small-object MOT. Full article

Remote towers are increasingly deployed at small-to-medium airports globally for cost efficiency, yet safety optimization for large airport remote apron control remains underexplored. This study proposes a human error identification framework for air traffic controllers (ATCOs) in large airport remote apron operations. Using hierarchical task analysis (HTA), a cognitive-behavioral model, and the technique for retrospective analysis of cognitive errors (TRACEr), we analyzed error probability and severity through field research. Key findings reveal critical divergences. Memory functions showed the highest error probability, while perception errors caused the most severe outcomes. Working memory errors were most prevalent, but visual detection errors were most severe. Attention deficits were most frequent, while spatial confusion and information integration failures exceeded severity thresholds. Personal factors dominated performance-shaping factors, with low vigilance and equipment unavailability as primary high-risk conditions. This research provides an error identification checklist and analysis methodology to enhance human performance and aviation safety in remote apron control. Full article

The escalating global surge in Waste Electrical and Electronic Equipment (WEEE) necessitates the strategic deployment of recycling facilities within resilient, multi-period networks. However, existing planning methodologies falter due to the non-stationary spatiotemporal volatility of e-waste generation, the high reconfiguration costs associated with path-dependent infrastructure, and the “curse of dimensionality” inherent in large-scale dynamic optimization. To address these challenges, we propose LAtt-PR, an innovative hybrid reinforced adaptive optimization framework. The methodology integrates a spatiotemporal attention-based neural network, combining Multi-Head Attention (MHA) for spatial correlation with Long Short-Term Memory (LSTM) units for temporal dependencies to accurately capture and predict fluctuating demand patterns. At its core, the framework employs Deep Reinforcement Learning (DRL) as a high-level action proposer to prune the expansive search space, followed by a Particle Swarm Optimization (PSO) module to perform intensive local refinement, ensuring both global strategic foresight and numerical precision. Experimental results on large-scale instances with 150 nodes demonstrate that LAtt-PR significantly outperforms state-of-the-art benchmarks. Specifically, the proposed framework achieves a solution quality improvement of 76% over traditional metaheuristics Genetic Algorithm (GA)/PSO and 55% over pure DRL baselines Deep Q-Network(DQN)/Proximal Policy Optimization (PPO). Furthermore, while maintaining a negligible optimality gap of less than 4% relative to the exact solver Gurobi, LAtt-PR reduces computational time to just 16% of the solver’s requirement. These findings confirm that LAtt-PR provides a robust, scalable, and efficient decision-making tool for optimizing resource circularity and environmental resilience in volatile, real-world recycling logistics. Full article

We present a conceptual and computational investigation of vacuum memory within a discrete toy-model framework. In this phenomenological approach, we introduce an effective memory field that records virtual events and nonlocal couplings on a lattice, without claiming to derive a fundamental new field of nature. Using a discrete toy model, we simulate memory formation via virtual events, nonlocal links, and black-hole-like information sinks. The resulting dynamics exhibit long-range spatial correlations, curvature-induced accumulation, high-entropy retention zones, and distinct spectral features, indicating that the modeled memory field can store and organize information in a vacuum-like medium. Building on this foundation, we incorporate curvature-modulated vacuum memory fields into Bohmian particle dynamics. By varying the memory coupling strength λ, we demonstrate that memory gradients systematically bend particle trajectories toward curvature centers, illustrating an active role for structured memory in guiding quantum-like motion. We further show that when vacuum memory encodes the full quantum phase S(x, t) and particles are guided by the Bohmian relation $\dot{x}=m^{-1}{\partial }_{x}S$, the trajectories collapse onto a single path with machine-level precision, providing a numerical consistency check that our implementation reproduces exact pilot-wave guidance and minimal-action dynamics. Through a minimal two-site entangled-memory model, we demonstrate that coupled memory fields—without explicit particle dynamics—can spontaneously synchronize via weak informational coupling, generating robust nonlocal correlations reminiscent of entanglement. Finally, we simulate two-slit interference under vacuum memory perturbations. While random, unstructured memory preserves quantum coherence and fringe visibility, structured, phase-sensitive memory induces dephasing and suppresses interference, functioning as a phenomenological decoherence mechanism. Together, these results situate our toy model within emerging information-based views of quantum dynamics and spacetime, offering a computational platform and conceptual lens for exploring the informational dynamics of a vacuum-like medium. Full article

Aging disrupts the neurovascular unit (NVU) and blood–brain barrier (BBB), elevates glial inflammatory tone, and compromises hippocampal memory. Environmental enrichment (EE)—a multimodal, lifestyle-based intervention—improves cognition, but its association with BBB/NVU and FNDC5/irisin-related signaling in aging remains incompletely understood. Aged male C57BL/6J mice (21 months old) were housed under EE or standard conditions for 11 weeks. Hippocampal-dependent spatial working memory was assessed using the radial eight-arm maze, and neuronal (NeuN), glial (Iba1, GFAP), and BBB/NVU markers (AQP4 endfoot polarity, occludin, ZO-1, PECAM-1, microvessel length/density) were quantified. FNDC5/irisin-related signaling was evaluated by measuring PGC-1α, FNDC5/irisin, IGF-1, BDNF, pAKT, and serum irisin. EE improved spatial working memory in aged mice, reducing working-memory errors, increasing correct choices before the first error, and enhancing path efficiency. EE attenuated the age-related decline of NeuN(+) neurons in the hippocampal CA1 and CA3 regions and suppressed microglial and astrocytic activation. EE strengthened BBB/NVU integrity by restoring AQP4 endfoot polarity, increasing occludin, ZO-1, and PECAM-1, and increasing cortical microvessel length and density. At the molecular level, EE upregulated the PGC-1α–FNDC5/irisin–IGF-1 axis and was accompanied by increased cortical BDNF and pAKT levels, as well as elevated circulating irisin, changes that occurred in parallel with NVU stabilization and reduced glial activation. EE mitigates age-related cognitive decline in association with coordinated neuronal, glial, vascular, and FNDC5/irisin-related signaling changes, supporting BBB/NVU preservation and cognitive resilience during aging. Full article

Locust outbreaks cause a significant threat to global food security and ecosystem stability, with particularly severe consequences in grassland regions, where grasshoppers also exert considerable ecological pressure. In comparison to grasshoppers, locusts typically occur at much larger spatial scales, as their strong migratory ability and collective movement behavior lead to greater spatial connectivity and autocorrelation. The forecasting of both locust and grasshopper outbreaks remains a formidable scientific challenge, primarily due to the complex, nonlinear spatiotemporal interactions among environmental drivers such as weather, vegetation, and soil conditions. This review compares the evolution of prediction methodologies for locust and grasshopper outbreaks, focusing on the application of deep learning (DL) methods to ecological forecasting tasks. It traces the development from traditional statistical models to classical machine learning, and ultimately to DL, assessing the strengths and limitations of key DL architectures—including Deep Neural Networks (DNNs), Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) networks, and Gated Recurrent Units (GRUs)—in modeling the intricate dynamics of locust populations. While most studies have concentrated on locust outbreaks, this review emphasizes the adaptation of these models to grassland ecosystems, such as those in Inner Mongolia, where grasshopper outbreaks exhibit similarities to locust plagues but have been largely overlooked in DL research. Despite the potential of DL, challenges such as data scarcity, limited model generalizability across regions, and the “black box” issue of low interpretability remain. To address these issues, we propose future research directions that integrate Explainable AI (XAI), transfer learning, and generative models like GANs to development more robust, transparent, and ecologically grounded forecasting tools. By promoting the use of efficient architectures like GRUs within customized frameworks, this review aims to guide the development of effective early warning systems for sustainable locust management in vulnerable grassland ecosystems. Full article

Seasonal groundwater (GW) pumping and climatic variability significantly impact the dynamics of greenhouse-dominated agricultural systems, yet quantitative evaluations at the local scale remain limited. This study explores non-parametric statistical and deep learning (DL) models for analyzing seasonal GW trends and predicting GW levels near greenhouse agriculture systems in Gyeongsangnam-do, South Korea. The modified Mann–Kendall (MK) test and Sen’s slope estimator were used to estimate long-term seasonal trends for the summer (wet season) and winter (dry season), based on monthly GW-level time series from six monitoring wells. Findings indicate that seasonal asymmetry is strong (winter trends have greater magnitudes and greater variability than summer trends), and that winter trends are negative (ranging from −0.45 to +1.70 m year⁻¹) and summer trends are positive (ranging from −0.02 to +0.31 m year⁻¹). At Jinju1 and Jinju4, statistically significant increasing trends were observed in both seasons (p < 0.05), but at other stations, weak or non-significant trends were observed due to short records or high variance. Long short-term memory (LSTM) and spatio-temporal graph neural network (STGNN) models were deployed and compared to predict at the GW level. The STGNN was found to be superior to LSTM in terms of R² (0.799–0.994) and reduced RMSE of up to 64.6, especially in winter, when spatially synchronized pumping is dominant in GW behavior. Despite advanced modeling, there is a serious concern about data limitations. Findings show that combining seasonal trend analysis with spatiotemporal modeling of DLs can significantly enhance knowledge and forecasting of GW dynamics in intensive greenhouse farming. Full article

Segmenting white matter tracts in diffusion-weighted magnetic resonance imaging (dMRI) is of vital importance for brain health analysis. It remains a challenging task due to the intersection and overlap of tracts (i.e., multiple tracts coexist in one voxel) and the data complexity of dMRI images (e.g., 4D high spatial resolution). Existing methods that demonstrate good performance implement direct volumetric tract segmentation by performing on individual 2D slices. However, this ignores 3D contextual information, requires additional post-processing, and struggles with the boundary handling of 3D volumes. Therefore, in this paper, we propose an efficient 3D direct volumetric segmentation method for segmenting white matter tracts. It has three key innovations. First, we propose to deeply interleave convolutions and transformer blocks into a U-shaped network, which effectively integrates their respective strengths to extract spatial contextual features and global long-distance dependencies for enhanced feature extraction. Second, we propose a novel channel-wise transformer, which integrates depth-wise separable convolution and compressed contextual feature-based channel-wise attention, effectively addressing the memory and computational challenges of 4D computing. Moreover, it helps to model global dependencies of contextual features and ensures each hierarchical layer focuses on complementary features. Third, we propose to train a fully symmetric network with gradually sized volumetric patches, which can solve the challenge of few 3D training samples and further reduce memory and computational costs. Experimental results on the largest publicly available tract-specific tractograms dataset demonstrate the superiority of the proposed method over the current state-of-the-art methods. Full article

Sixth-generation (6G) networks will operate with pervasive autonomy and minimal centralised control, imposing stringent requirements on security and trust. This short communication presents a hybrid trust evaluation approach that combines fuzzy inference for uncertainty management, bidirectional long short-term memory (BiLSTM) networks for temporal prediction, and blockchain for immutable verification. The pipeline first maps multi-source interaction and context metrics into linguistic trust values via fuzzy rules, then leverages BiLSTM to anticipate trust fluctuations under dynamic conditions, and finally anchors trust updates on a permissioned blockchain to ensure integrity and traceability. Using CIC-IoT2023, the proposed approach attains high accuracy and F1-score while reducing Execution Time (ET) and energy demands relative to a recent spatial-temporal trust model for 6G IoT. Results indicate that jointly addressing uncertainty, temporal evolution, and ledger-backed validation yields stable trust trajectories suitable for resource-constrained devices. The study outlines a practical path toward explainable, adaptive, and tamper-resistant trust management for 6G ecosystems. Full article

In this study, we aimed to investigate the cognitive and affective-emotional factors underlying math achievement in a sample of 169 Japanese elementary school children. Using structural equation modeling, we examined the contributions of fluid and crystallized intelligence, verbal and spatial working memory, and affective-emotional variables, including general anxiety, test anxiety, math anxiety, and math self-efficacy. We found intelligence to be a strong positive predictor of math achievement, while among the affective-emotional variables, math self-efficacy emerged as the only significant predictor of math achievement. Interestingly, intelligence mediated the association between affective-emotional factors, such as math anxiety and self-efficacy, highlighting its central role in children’s math achievement. These findings underscore the strong relationship between intelligence and self-efficacy in educational contexts, suggesting that self-efficacy is closely linked to cognitive abilities to support children’s success in math. Educational implications are discussed, emphasizing the need to strengthen math self-efficacy alongside cognitive abilities. Full article

Students with attention deficit hyperactivity disorder (ADHD) face a continuous challenge with their attention span, putting them at a greater risk of academic or psychological difficulties compared to their peers. Innovative communication technologies are demonstrating potential to address these attention-span concerns. Virtual Reality (VR) is one such example, and has the potential to address attention-span difficulties among ADHD students. Accordingly, this study presents an EEG-based multimodal sensing pipeline as a methodological contribution, focusing on sensor-based data acquisition, signal processing, and neurophysiological interpretation to assess attention in VR-based environments, simulating a university supply chain educational topic. Thus, in this paper, a sequential exploratory approach investigated how 35 participants experienced an interactive VR-learning-driven supply chain game. A Brain–Computer Interaction (BCI) sensor generated insights by quantitatively analysing electroencephalogram (EEG) data that were processed through the proposed pipeline and integrated with subjective measures to validate participant’s subjective feelings. These insights originated from questions during the experiment that followed the Spatial Presence and Technology Acceptance Model to form a multimodal assessment framework. Findings demonstrated that the experimental group experienced a higher improved attention, concentration, engagement, and focus levels compared to the control group. BCI results from the experimental group showed more dominant voltage potentials in the right frontal and prefrontal cortex of the brain in areas responsible for attention, memory, and decision-making. A high acceptance of the VR technology among neurodiverse students highlights the added benefits of multimodal learning assessment methods in an educational setting. Full article