Search Results (54,347)

Wildfires pose increasing environmental and socio-economic risks, particularly in climate-sensitive and tropical regions, necessitating reliable satellite-based monitoring and predictive frameworks. This study presents a comprehensive survey of satellite-derived wildfire indicators and spatiotemporal modeling approaches, covering their historical development, current methodologies, and emerging research directions. We review major active fire and hotspot datasets derived from MODIS, VIIRS, and related platforms, along with key environmental drivers such as vegetation indices, meteorological variables, and land-surface with a specific case study for the Indonesian region. Modeling approaches are synthesized from classical statistical regression and time-series analysis to contemporary machine learning and deep learning architectures, including convolutional neural networks, recurrent neural networks, and transformer-based models. The analysis highlights the transition toward multi-source data integration and spatiotemporal deep learning frameworks capable of capturing complex wildfire dynamics. Finally, we identify future research challenges, including hybrid physical–AI modeling, uncertainty quantification, and scalable real-time wildfire intelligence systems. This survey provides a structured reference for researchers and practitioners seeking to advance satellite-based wildfire monitoring and prediction. Full article

The Late Ediacaran Dengying Formation in the Sichuan Basin is a key target for deep-gas exploration, yet the controls on facies differentiation in the Penglai area, on the northern slope of the Central Sichuan Paleo-Uplift, remain debated. This study integrates 3D seismic data, well logs, cores, and thin sections from 20 wells to characterize the microfacies, microfacies associations, and sedimentary architecture of the Dengying Formation. Nine microfacies types are recognized and grouped into three microfacies associations: tidal flat, mound–shoal complex, and inter-mound. The same nine microfacies types occur in both members, but the mound–shoal complex association in the third and fourth members is more grain-rich, more strongly overprinted by recrystallization, and makes up meter-scale shallowing-upward cycles. The lateral variations in microfacies types across the study area are minor; facies differentiation is instead expressed through variations in microfacies associations and their cumulative thicknesses. Facies distribution analysis and 3D seismic paleogeomorphology demonstrate that the Penglai area developed a broad, gently NW-dipping carbonate ramp without distinct slope breaks, in contrast with the previously proposed rimmed platform model of the Gaoshiti–Moxi area. Within this ramp, mound–shoal complexes record large-scale lateral migration driven by high-frequency relative sea-level fluctuations, but their along-strike continuity is constrained by NE-trending basement faults oriented sub-perpendicular to the NW-trending rift axis. A hierarchical three-level control on facies differentiation is proposed: regional paleogeomorphology provides the first-order NW-dipping ramp framework; syn-sedimentary basement faults impose a second-order segmentation; and high-frequency sea-level fluctuations drive the third-order stacking patterns. This hierarchical mechanism refines the sedimentary model of the Dengying Formation and provides a predictive basis for delineating high-quality reservoir targets along the northern slope of the Central Sichuan Paleo-Uplift. Full article

k-entanglement, including entanglement relative to full separability and genuinely multipartite entanglement, is a fundamental quantum resource in multipartite quantum systems. Its identification and quantification play essential roles in quantum information processing, quantum simulation, and quantum metrology. However, the practical computation of rigorous k-entanglement measures remains highly challenging due to the need for high-dimensional optimization. In this work, we propose a machine-learning-based surrogate framework for approximating the witness-based k-entanglement measure $E_w^{(k,n)}$. The numerical evaluation of the computationally realized quantity ${\tilde{E}}_w^{(k,n)}\left(\rho \right)$ is reformulated as a supervised regression problem, where the input is the density matrix $\rho$ and the labels are obtained from finite witness databases. The framework combines multilayer perceptrons (MLPs), convolutional neural networks (CNNs), and light gradient boosting machine (LightGBM) through a stacking ensemble. Numerical experiments are performed for 3- and 4-qubit systems as representative demonstrations of the proposed workflow. The results show that the learned models achieve high predictive accuracy in terms of MAE, MSE, and $R^2$, while providing millisecond-level inference for single-state evaluation. Werner state tests serve as symmetric benchmark checks, and an additional four-qubit noisy circuit-generated state family, obtained from finite-depth circuit preparation followed by local amplitude-damping noise, is used as a structured physical test beyond random density matrices. Compared with the optimization-based evaluation, the trained surrogate model significantly reduces the computational time while maintaining accuracy within the tested system sizes and data distributions. These results show that the proposed framework provides an efficient numerical surrogate for rapid approximation of witness-based k-entanglement measures, while extensions to larger systems and experimental data require further validation. Full article

Offline action prediction does not by itself guarantee reliable closed-loop flight for unmanned aerial vehicle (UAV) vision-language-action (VLA) models. We study a controlled AirSim Blocks waypoint task using 100 expert episodes and 3385 RGB-D, instruction, state, and action records. Checkpoints are selected only on val-seen data, after which val-unseen is evaluated once. A 132,840-parameter policy reaches $0.9278\pm 0.0019$ final-test action accuracy across three training seeds, yet raw VLA control fails in closed loop. We therefore embed its action proposals in progress-aware hybrid execution with explicit recovery and near-goal precision. The strongest checkpoint reaches 59/60 goals, but crossing three independently trained checkpoints with three target seeds yields a more conservative 147/180 successes (81.7%) with zero recorded collisions and marked checkpoint sensitivity. Substantial overrides and fallback-tagged steps further show that the reported closed-loop outcomes are properties of the hybrid system, not of the learned policy alone. These findings are restricted to the controlled AirSim Blocks benchmark and do not demonstrate real-UAV deployment, sim-to-real transfer, or field robustness. Full article

Understanding olive tree metabolism and its interactions with biotic and abiotic factors is crucial for the sustainability and resilience of olive cultivation in a changing agroclimatic context. In response to biotic stress, plants activate complex signaling pathways that trigger the production of specialized metabolites, particularly volatile organic compounds (VOCs). This study investigates the volatolomic profile naturally emitted by whole olive fruits using an integrated metabolomic strategy that combines design of experiments (DoE), targeted and untargeted analyses, and multivariate statistics. Optimal headspace solid-phase microextraction (HS-SPME) conditions were established using 30 g of sample, a 50 °C extraction temperature, a 50 min extraction time, and a 3 min injection at 250 °C, identifying extraction time and temperature as the most critical factors influencing VOC recovery. The data demonstrated significant cultivar-dependent variation in the volatile emissions from healthy olive fruit among six representative varieties. Furthermore, robust partial least squares-discriminant analysis (PLS-DA) and random forest models provided a clear separation between healthy and damaged olive fruits, achieving high predictive accuracy (90%) and identifying key volatile biomarkers derived from the lipoxygenase (LOX) pathway. This novel multivariate optimization approach (SPME–GC/MS) represents a powerful tool for establishing a reliable chemical fingerprint of the olive fruit “volatilome” under evolving agroclimatic challenges. Full article

Background/Objectives: Some patients with severe hypoxic–ischemic brain injury after out-of-hospital cardiac arrest (OHCA) progress toward brain death, a trajectory not adequately captured by the conventional classification of favorable versus unfavorable neurological outcomes. We developed and internally evaluated a multimodal model combining quantitative brain computed tomography (CT), serum neuron-specific enolase (NSE) at 48 h, and clinical variables to predict operationally defined progression toward brain death (PTBD). Methods: This multicenter retrospective secondary analysis used prospectively collected data from the Korean Hypothermia Network registry. Adult comatose OHCA survivors treated with targeted temperature management between October 2015 and December 2020 were included. Multivariable logistic regression models were developed in the total cohort and in patients with poor neurological outcomes. Model performance was assessed using discrimination, calibration, the Brier score, and bootstrap internal validation. Results: Of 468 patients assessed, 376 were included; their mean age was 58.7 years, and 269 (71.5%) were male. Seventy-four patients (19.7%) met the operational definition of PTBD. In the total cohort, younger age, non-shockable rhythm, low gray-to-white matter ratio (GWR ≤ 1.19), and higher NSE at 48 h were independently associated with PTBD. Among 288 patients with poor neurological outcomes, younger age, low GWR, and higher NSE at 48 h remained independent predictors. The total-cohort model had an AUC of 0.895 and an optimism-corrected AUC of 0.890. Its AUC was higher than that of NSE at 48 h (p < 0.001) but not significantly different from that of GWR alone (p = 0.054). In the poor-outcome subgroup, the model had an AUC of 0.864 and an optimism-corrected AUC of 0.858 and significantly outperformed both GWR (p = 0.012) and NSE at 48 h (p < 0.001). Conclusions: PTBD represents a clinically distinguishable trajectory among patients with poor neurological outcomes after OHCA. A multimodal model using information available within 48 h demonstrated good internally validated performance and may support early risk stratification before definitive neuroprognostication. External validation is required before clinical implementation. Full article

The increasing demand for predictive maintenance in industrial environments requires edge-intelligent solutions in which latency, energy consumption, memory footprint, and data movement are strictly constrained. This paper presents a two-stage Edge AI architecture for rolling-bearing prognostics, designed for heterogeneous embedded deployment across two distinct processing levels: a smart programmable sensing unit for local low-complexity signal preprocessing and a low-power embedded multiprocessor for higher-level temporal prognostic reasoning. In the first stage, a tiny neural preprocessor processes high-frequency vibration measurements directly at the sensing level and produces a compact low-dimensional degradation representation, complemented by lightweight health-related physical features. In the second stage, an embedded temporal reasoning model analyzes sequences of these compressed representations to estimate bearing degradation and remaining useful life. In addition to numerical remaining useful life regression, the second stage includes diagnostic reasoning heads that produce maintenance-oriented categories from a restricted vocabulary, enabling interpretable diagnostic summaries without relying on cloud-based language models. The complete pipeline is designed for fully edge-resident operation and exported in deployment-compatible formats, with the objective of supporting practical integration into heterogeneous industrial edge platforms. The proposed framework is trained and evaluated on the PRONOSTIA bearing degradation dataset and positioned against representative recurrent and hybrid prognostic baselines from the literature. From the deployment viewpoint, the sensor-side stage requires 68.62 ms inference time with 2.07 KiB RAM and 1.35 KiB flash/weights, whereas the edge temporal stage runs in 49.2 ms with 90.68 MiB RAM and 67.71 MiB flash/weights. In terms of prognostic performance, the proposed model achieves an average normalized RMSE of 0.1616 and an average normalized MAE of 0.1311 on three held-out bearings, while the weakly supervised diagnostic heads reach accuracies of 0.9066 for degradation trend and 0.8872 for vibration evidence. Experimental results show that the proposed architecture provides an effective trade-off between compact sensor-side processing, temporal prognostic accuracy, monotonic degradation consistency, hardware deployability, and interpretable maintenance-oriented outputs, supporting the feasibility of fully edge-based predictive maintenance systems for rolling-bearing health monitoring. Full article

To accurately predict temperature rise due to the concrete hydration heat released from the #0 segment of a continuous concrete girder bridge at an early construction stage for structural design purposes, researchers proposed an approach incorporating empirical predictive formulae with a preliminary numerical analysis. However, due to the uniqueness of the structural geometry and material in each engineering case and the limited data shared by the whole engineering community, no universal predictive empirical model for temperature rise due to hydration heat has yet been identified for practical use that can be applied to a variety of different projects. Moreover, the preliminary numerical analyses are usually based on questionable assumptions and simplifications of the physical truth, the accuracy of which also requires further validation. To this end, the present research measured the time-varying temperature samples of early age massive concrete in the #0 segment of Huangsha Harbor Bridge (a twin-deck three-span continuous concrete box girder bridge located in Jiangsu Province, China) and examined the accuracy of the predictive empirical models formulated by other researchers and the usability of a numerical modal established on a commercial finite element (FE) platform by comparing the corresponding results with the data from the present field measurements. The results suggest that the empirical formulae proposed can generally effectively describe the actual temperature distribution patterns related to the thermal issue, but they are characterized by inferior usability in some cases. In addition, the present comparison also indicates that the actual maximum temperature rise can be correctly predicted by the preliminary FE analysis in most cases. Full article

Based on survey data from the National Ecological Civilization Pilot Zone (Jiangxi), this study divides rural residents’ self-reported policy compliance in household waste classification (PC) into habit-based policy compliance in waste classification (HPC) and decision-based policy compliance in waste classification (DPC). The integration of a random forest model with Shapley Additive Explanations (SHAP) is utilized to identify the key predictors of each compliance type and to reveal the non-linear predictive contributions and marginal effects through which these factors respectively affect HPC and DPC. Furthermore, fuzzy-set qualitative comparative analysis (fsQCA) is used to uncover the multiple configurational pathways driving each type of compliance. Research indicates the following: (1) The key predictive factors of HPC and DPC are different. (2) Subjective norms (SUN), information interaction (II), and social norms (SON) exhibit significant nonlinear predictive contributions to HPC; policy identification (PI), SON, and procedural fairness (PF) show nonlinear trends in their contributions to the predictive probability of DPC. (3) There are five configuration paths for HPC and four for DPC. This study not only deepens the understanding of rural residents’ HPC and DPC but also provides practical guidance for governments to formulate differentiated and efficient measures to encourage compliance with waste classification policies. Full article

While drone delivery has gained significant scholarly and industrial interest, rural residents’ acceptance of these systems remains underexplored, despite the region’s acute logistics challenges and service inequities. Using survey data from rural U.S. residents, this study first estimates an ordered logistic regression (OLR) model to identify factors associated with five-level drone delivery acceptance. The study then compares OLR, multinomial logistic regression (MNL), Random Forest (RF), XGBoost, and LightGBM under matched feature sets to evaluate whether nonlinear machine-learning models improve prediction beyond the interpretable statistical baseline. Open-ended responses are coded into LLM-derived sentiment labels and added as supplementary predictors to test whether unstructured feedback improves acceptance prediction. Results show that willingness to pay is the strongest predictor of acceptance, while equitable same-day delivery demand and post-pandemic attitude adjustment are also positively associated with higher acceptance. Household disability status and urban accessibility are not significant after adjustment. In the five-level analysis, OLR provides a strong ordinal baseline, while XGBoost and other tree-based models improve selected class-level prediction metrics. In the binary high-acceptance analysis, machine-learning models show stronger predictive performance, especially when structured predictors are combined with sentiment features. This study contributes to rural drone-delivery literature by linking service equity, perceived value, and LLM-derived sentiment within a comparable statistical and machine-learning framework for rural service design. Full article

To address the challenges of information fusion and prediction for highly non-stationary, noisy, and lag-responsive heterogeneous time-series data from multi-source environmental sensing, this paper proposes a novel adaptive hybrid framework, SAPSO-VMD-GRU. First, the framework employs Variational Mode Decomposition (VMD) to decouple the target sequence into trend, periodic, and random components to reduce complexity and filter noise. Then, Lagged Cross-Correlation Analysis (LCCA) is introduced to quantify the time-lagged correlation between the target components and variables such as rainfall and multi-depth soil temperature and moisture, eliminating redundant features to achieve deep fusion of multi-source information. This paper designs an adaptive particle swarm optimization algorithm, SAPSO, by integrating improved Circle chaotic initialization, Sa-function-based nonlinear inertia weight, and a two-stage Cauchy mutation strategy. SAPSO is used to adaptively determine the VMD parameters and the key GRU hyperparameters in different modeling stages. Experiments based on the Bayintun landslide dataset show that, by utilizing the past 5 days of multi-source historical data, including GNSS displacement, rainfall, and lagged soil moisture and temperature, as inputs to forecast the next-day displacement, the proposed framework achieved $R^2$ values above 0.95 on the chronological hold-out validation subset at all three GNSS monitoring stations. These results indicate that the proposed framework can effectively capture lagged triggering effects and improve displacement prediction accuracy under complex, noisy, and non-stationary monitoring conditions. Full article

Cementitious sand and gravel (CSG) is commonly used in construction engineering; however, its mix proportion design is complex, and traditional physical experiments face limitations such as long cycles, high costs, and susceptibility to external factors when obtaining high-quality sample data. In this study, a foundational dataset was first acquired through physical experiments: 100 sets of CSG specimens with different mix proportions (cement content 40, 50, 60, 70 kg/m³; water-to-binder ratio 1.0, 1.2, 1.4; sand ratio 0.1, 0.2, 0.3, 0.4; fly ash content 20, 30, 40, 50 kg/m³) were prepared. After 28 days of standard curing, compressive strength and splitting tensile strength tests were conducted using a WAW-1000 electro-hydraulic servo universal testing machine, yielding 100 sets of real mechanical property data. The coefficients of variation for all test groups were below 10%, confirming the reliability and repeatability of the experimental data. On this basis, a data augmentation method based on Conditional Tabular Generative Adversarial Networks (CTGAN) is proposed. Through adversarial training between the generator and the discriminator, the model learns the multi-dimensional distribution characteristics of the original CSG data and generates 100 synthetic samples, which are then merged with the original data to expand the dataset to 200 samples. The quality of the synthetic data is evaluated using Wasserstein distance and correlation matrix heatmaps. Furthermore, a hybrid XGBoost–LSTM prediction model is proposed—XGBoost is used for feature construction to capture nonlinear interactions among mix proportion variables, and the constructed features are then fed into an LSTM network for sequential learning and regression prediction. The results show that the CTGAN-generated data are highly consistent with the original data in terms of kernel density distributions and variable correlations, with Wasserstein distance significantly superior to four comparative methods: Bootstrap, SMOTE, GaussianCopula, and TVAE. After augmentation, the XGBoost–LSTM model achieves a coefficient of determination (R²) of 0.9897 for compressive strength prediction (vs. 0.9793 before augmentation) and 0.9801 for splitting tensile strength (vs. 0.9882 before augmentation, a slight decrease). The mean absolute percentage errors (MAPE) are 4.49% and 4.11%, and the root mean square errors (RMSE) are 0.201 and 0.049, respectively; both error metrics are reduced compared with those before augmentation. Compared with baseline models including XGBoost, LSTM, Random Forest (RF), and Support Vector Regression (SVR), the XGBoost–LSTM model exhibits the best performance across all evaluation metrics, and Wilcoxon signed-rank tests confirm that the performance differences are statistically significant (p < 0.05). The proposed method of CTGAN-based data augmentation combined with the XGBoost-LSTM hybrid model provides an effective solution to the problem of insufficient CSG sample data and offers a reference for data enhancement and performance prediction of other small-sample materials. Full article

Rapidly developing, mid-sized Canadian municipalities often suffer from a deficit in dedicated modernization capacity, leaving public infrastructure lagging behind growth and reliant on historically “grandfathered” legacy solutions. To overcome the lack of empirical, data-backed planning in these regions, this paper proposes an agile, data-driven smart city framework centered around a localized digital twin (DT) environment. The framework is evaluated through a case study of a proposed new public transit route in Abbotsford, British Columbia, a rapidly expanding city grappling with decentralized commercial zones and low-density sprawl. Our approach synthesizes heterogeneous, multi-source spatial data, including regional commuter trajectories, real-time Abbotsford International Airport (YXX) flight schedules, and points of interest (POI) business densities, to map high-resolution hourly temporal variations in traffic conditions. These streams feed into a virtual simulation framework that evaluates operational cost–benefit trade-offs for proposed transit routes. Crucially, this framework serves as a living, continuously updated system that enables resource-constrained cities to dynamically simulate transit networks as commercial footprints and transit volumes evolve. Finally, we discuss the roadmap for this framework, detailing how integrating predictive AI models and gamified interfaces can democratize urban planning, enabling municipal stakeholders and non-technical operators to interactively co-design public transit systems. Full article

Off-grid photovoltaic systems with battery storage are essential for sustainable rural electrification, yet national programs such as Ethiopia’s NEP 2.0 lack frameworks that support decentralized alternatives. This study introduces a novel swarm electrification model, in which higher-tier solar systems trade surplus energy to support lower-tier households, forming a peer-to-peer solar-sharing network. A comparative assessment of solar resources using models, predictions, and satellite databases showed stable annual irradiance in Ethiopia, ranging from 4.22 to 6.54 kWh/m²/day across two predictive models and two satellite datasets. Long-term PVGIS data (13-year average) recorded the highest annual value at 7.30 kWh/m²/day. Statistically, the artificial neural network yielded the lowest error margins, while the Allen Regression model offered the lowest bias. Based on these data, Tier 2 and Tier 3 PV systems were designed and simulated at 85% efficiency with three-day battery autonomy. A 400 Wp PV array paired with a 2 × 250 Ah battery bank was designed to meet the Tier 3 daily demand of 1.7 kWh, generating over 60% energy surplus. Peak consumption occurs during evening hours (17:00–19:00). Lithium-iron-phosphate batteries proved economically superior for Tier 3 loads exceeding 1.5 kWh/day over a 10–15-year lifecycle, requiring zero replacements and offering lower overall costs. The hierarchical tier-based model enables strategic cross-subsidization, where Tier 3 households support Tier 1 and Tier 2 users. A comparative cable topology analysis recommends the radial T3 2@12V configuration for linear households within 10 m, and the ring T3 topology for longer linear layouts of 15–25 m requiring moderate fault tolerance. All configurations maintain voltage drop below the critical 5% threshold. Overall, this study demonstrates that optimized off-grid PV systems with appropriate topology and battery selection offer a sustainable and scalable pathway for rural electrification in Ethiopia. Full article

Evaluating wearable assistive robots in real-world caregiving is challenging because temporally aligned and repeatable A/B comparisons are rarely available. Conventional evaluations assume that assist-on and assist-off trials are comparable in task content, posture, and movement timing. However, caregivers adjust their posture and timing across human–human interactions and task sequences. This study proposes a causally inspired diagnostic framework based on DBN/SCM-inspired time-series modeling and movement-fixed counterfactual estimation. We represented the multimodal observations using intervention, robot state, movement context, EMG, and context variables. Node-specific relationships were approximated using Attention-based Sparse Variational Gaussian Process regressors. We evaluated the framework at three levels of environmental complexity. These levels comprised controlled trunk flexion, partially controlled bed-to-wheelchair transfer, and real-world caregiving. The proposed framework is intended as a diagnostic counterfactual evaluation tool rather than as a method for strict causal identification. Across experiments, one-step EMG prediction accuracy alone was insufficient to identify intervention-sensitive models. In controlled validation, the selected movement-decoupled robot-only model reproduced an EMG-reducing response consistent with the controlled A/B reference. When fitted to the partially controlled transfer data, the selected structural specification identified an EMG-increasing response in supported contexts. In the real-world caregiving case study, the global assist-mediated response (AMR) was near zero despite a positive pooled A/B difference. However, the stratified analysis identified localized supported responses. The near-zero AMR indicates that the pooled difference was not reproduced through the modeled assist intervention–robot state–EMG pathway under fixed movement context. This result should not be interpreted as evidence of overall device ineffectiveness. These findings suggest that context-fixed counterfactual diagnosis can help interpret assistive responses under increasing environmental complexity. Full article