Search Results (692)

Accurate short-term load forecasting at disaggregated levels is critical for energy management in microgrids and institutional environments, yet it remains a challenge due to high consumption variability and limited contextual information. This paper proposes a hybrid model that combines Fuzzy ARTMAP neural networks with K-means clustering to improve hourly load forecasting using real data from a university microgrid. The methodology includes key preprocessing steps such as filtering low-load records, removing holidays, interpolating missing values, and applying cyclic encoding to standardize the data into 96 time intervals per day (15-min resolution). For each prediction, the average load profile of the five most recent weekdays is computed and compared to cluster centroids to identify the most similar group, which is then used to train the neural network. Results demonstrate consistent improvements in MAPE, RMSE, and MAE compared to the non-clustered baseline. The model showed robustness to non-stationary behavior and atypical patterns, even when relying solely on timestamp and load data. The proposed strategy outperformed conventional approaches and proved suitable for complex, data-limited environments. Full article

Background: Ultraviolet (UV) radiation is a major risk factor for skin cancer. Nevertheless, many people tan in the sun and in tanning beds. The aim of this study was to explore the reasons behind these behaviors and to investigate whether the reasons for tanning in the sun and in tanning beds differ. Methods: We used data from a nationwide survey study conducted in Germany, which included n = 4156 individuals aged 16 to 65 years. We assessed different reasons for outdoor and indoor tanning, as well as sociodemographic characteristics, skin type, and tanning behaviors. Results: While both outdoor and indoor tanners frequently reported relaxation, as well as feeling of light and warmth as reasons, outdoor tanners placed a greater emphasis on increasing vitamin D levels and health benefits. In contrast, indoor tanners were more focused on enhancing attractiveness and pre-tanning for holidays. Individuals who sought a tan more frequently—whether through indoor or outdoor tanning—were more likely to agree with the various reasons provided, compared to those who tanned less often. In addition, we found associations with sex, age, immigrant background, education, occupation, and skin type. Conclusions: There were differences as well as similarities in the reasons for indoor and outdoor tanning. This indicates that overarching prevention strategies could be effective. Additionally, targeted measures specifically for indoor and outdoor tanning could also be beneficial in raising awareness about the risks of UV radiation and, in the long term, reducing the incidence of skin cancer. Full article

Recreational vitality is widely recognized as a core metric for assessing the quality of human settlements. Elucidating the relationship between recreational vitality and landscape characteristics is crucial for guiding the optimization and quality enhancement of urban waterfront spaces. This study takes the micro-scale waterfront space of the Beijing–Hangzhou Grand Canal (Hangzhou section) as its research object, systematically analyzes the correlation between waterfront landscape attributes and recreational vitality, and formulates specific optimization strategies for enhancing recreational vitality. A total of 310 representative sampling sites was established. The study integrates machine learning-driven semantic image segmentation to achieve refined quantification of waterfront landscape metrics and employs anonymized mobile phone signaling data to dynamically characterize the spatiotemporal distribution of recreational vitality. Through correlation analysis and regression modeling, it quantifies the effect size and functional mechanisms of key landscape metrics on recreational vitality, and further proposes adaptive strategies for recreational vitality enhancement tailored to different urban functional zones. The key findings are as follows: (1) Recreational vitality is significantly higher on holidays than on workdays. High-vitality areas are concentrated in commercial functional zones, with an overall spatial gradient of “low in the east and high in the west, low in the north and high in the south”. (2) High-level Green View Factor (HGVF) shows a stable positive correlation with vitality, whereas the Sky View Factor (SVF) and the Enclosure Interface View Factor (EIVF) correlate negatively. (3) The influence of landscape metrics is strongly moderated by functional zone type: in residential functional zones, HGVF has strong explanatory power; in commercial functional zones, it shows complex nonlinearity; in ecological conservation zones, its explanatory power is generally weaker. (4) Tailored enhancement strategies are proposed for each functional zone. This study clarifies the link between core waterfront landscape attributes and micro-scale recreational vitality, and provides a scientific basis for evidence-based design and sustainable enhancement of urban waterfront spaces. Full article

Background/Objective: Emergency Departments (EDs) face persistent challenges with overcrowding, unpredictable patient arrivals, and difficulty forecasting short-term demand. Precise hourly arrival predictions are crucial for effective staffing, optimal resource management, and minimizing entry delays. Methods: This paper develops and evaluates a forecasting framework comparing six approaches (a Seasonal Naive baseline, Exponential Smoothing (ETS), Ridge Regression, LightGBM, a hybrid Temporal Convolutional Network (TCN), and a hybrid Long Short-Term Memory (LSTM) network) using de-identified hourly patient arrival records from an ED in Madinah, Saudi Arabia, covering January–November 2024. A set of 183 engineered features is constructed from cyclical time encodings, weekend and public-holiday indicators, structured autoregressive lags, and volatility measures, with all lag-based features verified to use strictly retrospective information. Models are optimized using Bayesian hyperparameter search and trained under an asymmetric loss function that penalizes underprediction to reflect operational risk. Results: Results on a 14-day hold-out test set show that Ridge Regression achieves the lowest MAE (3.75, R² = 0.52), with TCN and LSTM essentially tied (MAE 3.80 and 3.85). Diebold–Mariano tests confirm that Ridge, TCN, and LSTM are statistically indistinguishable from one another and that Ridge is marginally significantly better than LightGBM ($p=0.028$); all four ML models significantly outperform ETS and the Seasonal Naive baseline ($p<0.001$). On the asymmetric metric, TCN achieves the best AsymRMSE (5.59), reflecting its tendency to err on the safe side of staffing decisions. Robustness is confirmed through sensitivity analysis across penalty factors, feature ablation demonstrating the contribution of each feature group without overfitting, expanding-window cross-validation across three independent monthly test periods, and conformal prediction intervals achieving well-calibrated coverage. Conclusions: These results demonstrate that combining engineered temporal features with either a lightweight linear model or a hybrid sequence model yields accurate hourly ED arrival forecasts; whether the achieved accuracy is operationally sufficient for staffing decisions remains a site-specific question that requires clinical validation beyond the scope of this single-center study. Full article

Electrical load time series exhibit strong heterogeneity across daily patterns driven by calendar effects and behavioral variability, leading many forecasting models to favor dominant weekday profiles while degrading on weekends, holidays, and transition days. This paper proposes an imbalance-aware spatiotemporal forecasting framework via a cluster-conditioned state space model. Daily load patterns are identified via time-series clustering and incorporated as conditioning covariates within a sequence-continuous selective state space models (Mamba), preserving temporal coherence without explicit sequence partitioning. A cluster-weighted training objective further mitigates pattern imbalance while avoiding future-information leakage. The resulting cluster-conditioned Time Series Mamba (TSMamba) consistently improves forecasting robustness across both frequent and infrequent profiles, achieving weighted absolute percentage error (WAPE) reductions of approximately 15% on weekdays, 42% on weekends, and 39% on holidays relative to the vanilla TSMamba, with similar gains in mean absolute error (MAE) and coefficient of variation of the root mean square error (CVRMSE). These results demonstrate that conditioning state dynamics on latent load patterns yields stable and computationally efficient short-term load forecasts under profile transitions. Full article

The annual occupational personal ultraviolet radiation (UVR) exposure of outdoor workers is vital for several purposes, including non-melanoma skin cancer risk assessment and the recognition of UVR-related pathologies as occupational diseases. Estimations of annual personal exposure (PE) are based on measurements, which are influenced by the measuring period respectively by the start and end time of the measurements, and PEs gained from different periods may differ noticeably. Therefore, we present a method that recalculates PE measurements to any other period (time and duration) during the day, and which is also applicable for measured ambient UVR to determine the relative personal UVR exposure (ERTA). The application shows the necessity of considering not only duration but especially time, as noon hours contribute differently than morning and evening hours. The uncertainties of recalculations are within ±5% if the measuring or target periods last at least 5 h and noon hours are covered. Furthermore, we propose a method to calculate annual PE using ERTA. The application for Austria shows that depending on the work-time model (working hours, working times) and date of holidays, annual PE may differ by up to 30%. Additionally, interannual variability of 16% within a ten-year period suggests avoiding a single year for consideration. Full article

Against the backdrop of smart cities and digital cultural tourism, the accurate prediction of urban passenger flow is of great significance for public security management and resource allocation. However, existing studies mostly rely on single data sources or only perform a simple concatenation of multi-source features, lacking systematic indicator system design. Meanwhile, weekly or monthly data are commonly used with coarse temporal granularity, making it difficult to capture short-term fluctuations and lag effects. To overcome these limitations, this paper collects the daily passenger flow data of Hangzhou from 15 March 2024 to 15 March 2025; integrates multi-dimensional factors such as keyword search trends across platforms, holidays and major events, and online public opinion; and constructs three daily characteristic indicators: online search index, humanistic–meteorological index, and textual sentiment index. The data denoising, dimensionality reduction, and sentiment quantification are realized through methods including SSA, PCA, and SnowNLP. On this basis, a hybrid CNN-GRU model integrated with the attention mechanism is proposed. An improved artificial bee colony (ABC) algorithm is adopted for global hyperparameter optimization, and a weighted hybrid loss function (JQHL) is introduced to enhance the model’s adaptability to extreme values. The results show that the ABC-CNN-GRU-Attention model, incorporating multi-dimensional indicators, outperforms traditional methods on evaluation metrics, including MAE, RMSE, MAPE, R², and RPD, demonstrating a higher prediction accuracy and robustness. Full article

Background and Objectives: Combining osimertinib (OSI) with pemetrexed (PEM) can enhance antitumor efficacy; however, the benefit is schedule-dependent. Our previous pharmacodynamic (PD) study showed that concurrent PEM + OSI is limited by OSI-induced G1 arrest, attenuating early PEM cytotoxicity. In contrast, sequential PEM→OSI allows PEM to fully induce S-phase arrest and DNA damage but elicits a pro-survival EGFR rebound; subsequent OSI suppresses this rebound and promotes apoptosis of damaged cells, yielding strong synergy. Here, we investigated whether pharmacokinetic (PK) drug–drug interactions (DDIs) contribute to this synergy and predicted the relative advantage of PEM→OSI versus PEM + OSI under clinically relevant conditions using a PK/PD approach. Method and Results: Potential PK-DDIs were assessed at cellular uptake, plasma exposure, and intratumoral distribution levels. No meaningful PK-DDIs were observed, supporting a primary PD-driven synergy. We integrated mouse PK with in vitro/in vivo PD data to build a mechanistic Quantitative System Pharmacology (QSP)–PK–PD model linking drug disposition to folate biology, Epidermal Growth Factor Receptor (EGFR) signaling, and tumor growth inhibition. The model recapitulated schedule dependence and explained PEM→OSI superiority: PEM initiates damage and EGFR compensatory rebound, after which OSI suppresses EGFR signaling and enhances apoptosis. In contrast, concurrent PEM + OSI induced G1 arrest, reduced the pool of damaged apoptosis-susceptible cells, and weakened the synergy. Global sensitivity analysis identified intrinsic OSI sensitivity and the pro-apoptotic protein Bim as key determinants; reduced OSI sensitivity or Bim activity diminished the advantage of the sequential strategy. The simulations indicated that OSI can start 48 h after PEM exposure (no extended drug holiday is needed) and that the PEM→OSI benefit remains robust across heterogeneity, including BIM-deletion polymorphisms and inter-individual variability in tumor drug sensitivity. Conclusions: This mechanism-based QSP–PK–PD framework connects whole-body PK to core PD processes, explains schedule-dependent synergy, and supports optimization of sequencing intervals and identification of likely responders. Full article

Accurately assessing the accessibility of fire services is critical for enhancing urban safety and the resilience of the built environment. However, existing studies often lack a systematic analysis of spatiotemporal dynamics across an entire municipality. To address this gap, this study develops a citywide dynamic assessment framework for Shanghai, integrating GIS with real-time traffic data across 240 consecutive intervals to assess the service accessibility of 195 fire stations in relation to 7973 key units of fire safety. The principal findings are threefold. First, the results reveal significant urban–suburban heterogeneity in emergency response times. Notably, the proximity advantage of fire stations in central urban areas is offset by traffic congestion, and the marginal benefit of traffic speed improvement exhibits a sharp decline once the average speed exceeds a critical threshold of 13.7–21.0 km/h. Second, the accessibility ratio demonstrates a clear temporal pattern, being highest on holidays and lowest during weekday peak hours, and follows a nonlinear spatial decline from the urban centre to the periphery. This pattern is influenced more critically by the matching of supply and demand than by fire station density alone. Third, the analysis identifies dynamic vulnerability hotspots, which display a ‘bimodal (M-shaped)’ pattern on weekdays and a ‘unimodal (A-shaped)’ pattern on weekends and holidays. This spatiotemporal mismatch shows that central urban areas, despite higher station density, can suffer from both high fire risk and low accessibility, revealing structural patterns consistent with the ‘Inverse Care Law’ in emergency service provision. This study concludes that merely improving traffic conditions is insufficient; optimising the spatial matching of resources is paramount for effective urban disaster prevention. By developing a refined dynamic assessment framework, this study advances current knowledge by focusing on demand locations consistent with actual fire regulatory priorities and examining spatiotemporal patterns across both urban and suburban areas, thereby providing quantitative, evidence-based support for the strategic planning of fire stations and the enhancement of infrastructure resilience. Full article

Background and Objectives: Weight gain during winter holidays has been reported in several studies, but most of them focus on short-term changes and rely primarily on body weight or BMI. This study investigates if body composition alterations associated with winter holiday period persist over a one-year follow-up in a Romanian adult population. Materials and Methods: This prospective longitudinal observational study included three assessment points: before the winter holidays (T1), immediately after the holidays (T2), and one year later (T3). Bioelectrical impedance analysis was used to obtain body composition parameters. A total of 120 participants completed all three assessments and were included in the longitudinal analysis. Results: Body weight and visceral fat area increased modestly yet significantly between T1 and T2. At the one-year follow-up, values remained similar to those observed immediately after the holiday period, suggesting persistence at group level (body weight: 68.25 → 69.40 → 69.45 kg and visceral fat area: 98.49 → 100.54 → 101.42 cm²). The net change observed during the holiday period was similar in magnitude to the overall annual difference. Changes in body weight were significantly associated with changes in visceral fat both during the holiday period and across the entire follow-up. Conclusions: Modest increases in body weight and visceral fat observed during the winter holidays were still present at one year. These findings suggest that short seasonal periods may contribute to overall annual changes in body weight and fat distribution. Full article

The vitality dynamics of historic urban areas under high tourism pressure and their underlying mechanisms remain not fully understood, posing a challenge to sustaining their uniqueness against homogenized redevelopment. To explore this issue, this study utilises human mobility data and an XGBoost-SHAP model to examine the spatio-temporal dynamics of block-level vitality and to uncover the nonlinear effects of built environment factors in Shenyang, China. The results indicate that: (1) Diverging from the commuting patterns of general urban areas, the vitality of historic urban areas presents unique spatio-temporal shifts, transitioning from commercial centers on weekdays to a commercial-cultural mix during holidays. (2) The determinants of vitality vary temporally, shifting from accessibility-oriented (subway) on weekdays to heritage-oriented (state historic sites) during holidays. (3) By applying the ‘Three-Factor Theory’ from satisfaction research to decode nonlinear effects, the study classifies factors into Performance (functional density), Basic (proximity to water bodies), and Excitement (distance to subway and state historic sites). The findings guide urban renewal to prioritize systemic and sustainable vitality across the historic urban areas rather than maximizing vitality in specific locations. Full article

Accurate one-step forecasting of scenic-spot visitor demand is challenging due to strong non-stationarity, holiday-induced peaks, and abrupt reputation-driven shocks. We propose a symmetry-aware dynamic graph learning framework that fuses social–physical sensing streams for robust demand prediction. Online reviews are treated as social sensing, transformed into daily sentiment indicators, and aligned with demand using a delay-aware aggregation scheme. To capture evolving inter-spot dependencies, we construct a time-varying adjacency matrix that is updated over time and integrated into a lightweight spatio-temporal forecasting model, Dynamic Spatio-temporal Graph Attention LSTM (DSGAT-LSTM). The model preserves the permutation-invariant property of graph learning while introducing sentiment-guided feature reweighting and sentiment-gated temporal updates to better track volatility. Experiments on multi-year daily data from multiple A-level scenic spots with holiday and weather context demonstrate consistent error reductions over representative temporal and graph-based baselines, together with improved stability under peak and shock conditions. We will release the processed feature-level dataset and implementation scripts to support reproducibility. Full article

Accurate forecasting of lottery sales is crucial for strategic planning in volatile consumer markets driven by trend shifts, multi-scale seasonality, and calendar effects. This study proposes a Diagnostic Decomposition–Correction Hybrid (DDC-Hybrid) framework integrating Prophet and SARIMA through a residual diagnostics and correction pipeline. Specifically, Prophet is employed to model long-term trend changes and interpretable holiday impacts, while SARIMA is subsequently used to correct the residual series, capturing short-range temporal dependence that remains statistically significant after decomposition. From an information-theoretic perspective, the framework can be viewed as a two-stage uncertainty reduction process, where decomposition extracts low-frequency informative components and residual correction harvests remaining predictive information. Using monthly lottery sales in China (2008–2025), we conduct a comprehensive evaluation of SARIMA, Prophet, and the proposed hybrid approach. The DDC-Hybrid demonstrates improved predictive accuracy, yielding the lowest error rates. Beyond predictive accuracy, we further examine varying holiday effects through statistical testing. We also find that lottery sales contain a pronounced quadrennial (48-month) seasonal cycle associated with mega-sport events, which improves long-horizon stability. The results suggest that the proposed diagnostic hybrid modeling approach enhances forecasting accuracy and provides practical insights for lottery sales management. Full article

To address the limitations of existing annual load scenario generation methods, including insufficient ability to represent long-term trends, excessive randomness in generated scenarios, and inadequate consideration of special holiday conditions, in this paper, an annual load curve generation method is proposed that integrates Seasonal–Trend decomposition using Loess (STL) with an improved denoising diffusion probabilistic model (DDPM). In the proposed method, the STL algorithm is first applied to decompose the annual load curve into a trend component and a daily seasonal component. The trend component is used as a baseline to ensure that the generated load curves remain consistent with the actual long-term trend characteristics. On this basis, an improved diffusion-based denoising model is employed to achieve controllable generation of different types of daily load scenarios. Finally, the generated daily load scenarios are aggregated with the trend component on an hourly basis to construct annual load scenario curves that simultaneously preserve realistic trend behavior and stochastic fluctuations. A case study based on a city in China is used to evaluate the proposed method. The results demonstrate that both the generated daily load scenarios and annual load scenarios outperform existing benchmark methods across multiple quantitative evaluation metrics, thereby validating the effectiveness of the proposed load scenario generation approach. Full article