Search Results (3,673)

Broken gangue in goafs exhibits complex mechanical deformation and seepage evolution under coupled loading and hydraulic action, which directly affects the hydraulic stability and water-hazard prevention of mining engineering. In this study, a systematic investigation was carried out to elucidate the evolution of seepage characteristics in a granular broken-rock assemblage under coupled hydraulic–mechanical loading. Four mono-sized specimen groups with particle-size ranges of 5–10 mm, 10–15 mm, 15–20 mm, and 20–25 mm were prepared. Using a modified rock triaxial–hydraulic testing system, nominal uniaxial compression tests, triaxial compression tests under different moisture conditions, and staged axial loading–seepage coupling tests were conducted. The results indicated pronounced particle-size effects: with increasing particle size, the nominal uniaxial compressive strength decreased (maximum reduction of 41.26%), while the crushing ratio increased (from 0.99% to 28.89%). The compression–densification process exhibited a staged evolution characterized by “slow increase–rapid increase–stable increase.” Water-induced deterioration intensified with increasing water content, and the compressive strength reduction reached 29.8% under saturated conditions. The evolution of seepage behavior was jointly governed by loading rate and particle size. Both pore pressure and pore-pressure gradient increased with loading rate. The permeability–porosity relationship was nonmonotonic, with an inflection occurring at a porosity of approximately 0.30–0.32, accompanied by an order-of-magnitude variation in the Darcy-flow deviation factor, indicating a progressive nonlinear deviation from Darcy behavior. These observations reflected a competitive mechanism involving “compaction-induced flow resistance increase–fragmentation and rearrangement–local channel regeneration.” Numerical simulations performed in COMSOL6.2 further confirmed, at the microscopic level, that the development of preferential local seepage channels and the expansion of stagnant-water zones were the fundamental causes of locally enhanced seepage capacity under an overall compaction background. The findings provide a theoretical basis for understanding water–rock interaction mechanisms in goafs and offer reference for mine water-hazard mitigation and groundwater resource protection. Full article

The Brazilian craft beer market has experienced continuous growth, increasing operational challenges for small- and medium-sized breweries that frequently rely on empirical and spreadsheet-based production routines. These practices often lead to inefficient resource allocation, production instability, and sustainability concerns. This study proposes an integrated analytical framework combining Machine Learning (ML) and Operations Research (OR) to improve demand forecasting and production planning. The methodology is based on a synthetic dataset calibrated to the operational conditions of a Brasília-based craft brewery, incorporating realistic demand patterns such as seasonality, trend, and intermittency across multiple SKUs over an 18-month horizon. Forecasting models—including Moving Average, Single Exponential Smoothing, and a global ML-based proxy—were evaluated using rolling-origin validation. The resulting probabilistic forecasts were integrated into a capacity-constrained optimization model based on linear programming, extended with risk-aware decision-making using Conditional Value-at-Risk (CVaR). The results indicate that the ML-based approach achieved competitive forecasting performance (sMAPE = 5.83% and MAE = 11.76) while enabling the generation of capacity-feasible and risk-aware production plans aligned with service-level targets. The integration of probabilistic forecasts into the optimization model allowed explicit trade-offs between cost, service level, and resource utilization. The main contribution of this study lies in demonstrating how the integration of predictive and prescriptive analytics can support more sustainable production planning in resource-constrained manufacturing environments. By replacing ad hoc spreadsheet routines with a closed-loop decision-support system, the proposed framework advances the literature on data-driven PPC and provides practical guidance for SMEs operating under uncertainty. Full article

This paper introduces the Polynomial Perceptron (PP), a structured extension of the classical perceptron that incorporates explicit polynomial feature expansions to model nonlinear interactions while preserving analytical transparency. By expressing feature interactions in closed functional form, PP captures higher-order dependencies through a compact set of learned coefficients, establishing a principled trade-off between expressivity and parameter efficiency. The proposed architecture is evaluated across heterogeneous domains, including text, image, and structured data tasks, under controlled experimental settings with parameter-matched baselines. Performance is assessed using standard metrics such as classification accuracy and model complexity (parameter count). Empirical results demonstrate that low-degree PP models achieve competitive accuracy compared to multilayer perceptrons and convolutional neural networks, while requiring significantly fewer parameters. An ablation study further analyzes the impact of polynomial degree on predictive performance, revealing diminishing returns beyond moderate degrees and highlighting favorable efficiency–accuracy trade-offs. A key advantage of PP lies in its intrinsic interpretability. Unlike conventional deep learning models that rely on post hhoc explanation methods, PP provides direct analytical insight through its explicit polynomial structure, enabling decomposition of predictions into feature-, token-, or patch-level contributions without surrogate approximations. Overall, the results indicate that PP offers a lightweight, interpretable, and computationally efficient alternative to standard neural architectures, particularly well-suited for resource-constrained environments and applications where transparency is critical. Full article

To address load–energy dynamic coupling in heterogeneous unmanned aerial vehicle (UAV) emergency rescue, this paper proposes an energy-coupled heterogeneous UAV task allocation (EC-HUTA) model that explicitly characterizes nonlinear interdependencies among payload, velocity, and power consumption, minimizing aggregate mission costs subject to physical and temporal constraints. To tackle the resulting high-dimensional, nonconvex problem, we introduce a multi-strategy improved stellar oscillation optimizer (MISOO), establishing a closed-loop synergistic system through three coupled stages: (i) evolutionary game-theoretic strategy competition via replicator dynamics for adaptive exploration–exploitation balance; (ii) intuitionistic fuzzy entropy (IFE)-driven dimension-wise parameter control, where IFE calibrates global exploration intensity while dimension-specific crossover probabilities accommodate heterogeneous convergence; and (iii) memory-driven differential escape mechanisms modulated by historical memory parameters to evade local optima. Cross-stage coupling through IFE ensures state information flows across the “strategy selection-refined search-dynamic escape” pipeline. Coupled with a dual-layer encoding scheme, this framework ensures efficient feasible search. Ablation studies validate each mechanism’s contribution. Evaluations on CEC2017 benchmarks demonstrate MISOO’s superior convergence against six metaheuristics. Large-scale earthquake rescue simulations confirm that EC-HUTA/MISOO strictly adheres to nonlinear energy constraints while enhancing task completion and temporal compliance. These results validate the framework’s efficacy for time-critical emergency resource allocation. Full article

This study develops a Life-Cycle Sustainability Framework for circular business models in the context of post-war economic reconstruction and sustainable value chain transformation. Ukraine is used as the main case study due to its post-war reconstruction context and the need for resource-efficient economic recovery strategies. Under conditions of disrupted supply systems, resource constraints, and structural economic change, circular economy principles are conceptualized as strategic mechanisms for enhancing resilience, resource efficiency, and long-term competitiveness rather than solely as environmental policy instruments. Building on a structured hierarchy of circular business models aligned with product life-cycle stages, the framework emphasizes value retention through functional and usage extension beyond material recovery. The framework includes a hierarchical classification of 12 circular business models and a sustainability evaluation approach based on four criteria (K1–K4), which allow for the comparative assessment of circular business models and their combinations across life-cycle stages. Using secondary statistical data and policy review as analytical inputs, the study identifies sectors with high potential for circular transformation and sustainable investment, including agriculture, energy, industry, construction, and logistics. The results indicate that circular business models applied at early life-cycle stages, such as reuse, repair, and remanufacturing, provide the highest potential for reducing resource intensity and improving long-term economic sustainability, while recycling and energy recovery play a supporting role. These findings highlight how life-cycle-oriented circular strategies can support sustainable reconstruction pathways, strengthen international cooperation, and inform policy and managerial decision-making in transitional economic contexts. Full article

A symbiotic mechanism exists between household livelihood capital and land use transition (LUT). However, previous studies have seldom examined, from a macro perspective, the characteristics of household livelihood capital and its impact on LUT, lacking an in-depth analysis of the spatial spillover effects and regional heterogeneity. Household livelihood capital influences LUT by shaping livelihood strategies. This article incorporates psychological capital into the Sustainable Livelihood Framework (SLF) and utilizes data from the China Family Panel Studies (CFPS) database and land use data spanning 24 provincial-level administrative units in China from 2010 to 2022; this study employs the Spatial Durbin Model to empirically analyze the impact of household livelihood capital on LUT and its regional variability, along with the spatiotemporal evolution patterns of household livelihood capital and land use change. The results demonstrate the following: (1) From 2010 to 2022, household livelihood capital increased, with higher levels of psychological, natural, and human capital, while social, financial, and physical capital were lower. The eastern region exhibited higher livelihood capital levels than the central and western regions, but the gap narrowed annually. (2) Between 2010 and 2022, LUT primarily involved the transition of cropland to construction land, cropland to forest and grassland, and forest and grassland to cropland. The intensity of LUT increased over time and showed a spatial gradient from west to east. (3) LUT positively affected the LUT of adjacent provinces. Various forms of capital and livelihood strategies had different effects on LUT within a province and across neighboring provinces. (4) The heterogeneity analysis reveals that the eastern region is significantly affected by the competitive effect and the spillover effect from neighboring provinces, while the central and western regions are constrained by water resources and inhibited by pure agricultural farmers. The article reveals the mechanism through which household livelihood capital drives LUT and the spatial spillover patterns, providing scientific evidence for regionally differentiated land management and ecological compensation policies, which is of great importance for the sustainable use of regional land resources. Full article

The Mediterranean Sea is a major biodiversity hotspot increasingly affected by biological invasions, climate warming, and habitat degradation. Among the most successful invaders are the rabbitfish species Siganus luridus and Siganus rivulatus, Lessepsian migrants from the Red Sea that are now widespread across the eastern Mediterranean. This study examined how these invasive herbivores influence native herbivore assemblages in shallow coastal habitats around Lipsi Island in the Aegean Sea, Greece. Using Underwater Visual Census (UVC) surveys and in situ feeding observations, we quantified the abundance and grazing activity of invasive rabbitfish relative to that of the native herbivores Sparisoma cretense and Sarpa salpa. Invasive rabbitfish represented approximately 35% of the herbivore assemblages and showed clear habitat and dietary preferences. Significant negative correlations were observed between invasive foraging activity and the feeding rate of the native S. cretense, while no such effect was found for S. salpa. High habitat overlap between S. luridus and S. cretense suggests that this native species may be more susceptible to competition on rocky substrates. Evidence of partial resource partitioning was observed, including increased use of seagrass habitats by S. salpa. These findings highlight how invasive herbivores can restructure native herbivore communities and alter grazing dynamics in eastern Aegean coastal ecosystems. Given the ongoing sea warming and widespread decline of seagrass habitats across the Mediterranean, understanding these competitive interactions is therefore essential for assessing future biodiversity trajectories and informing management strategies. Full article

Breast cancer histopathology classification remains a fundamental challenge in computational pathology due to variations in tissue morphology across magnification levels. Convolutional neural networks (CNNs) have long been the standard for image-based diagnosis, yet recent advances in vision-language models (VLMs) suggest they may provide strong and transferable representations for complex medical images. In this study, we present a systematic comparison between CNN baselines and large VLMs—Qwen2 and SmolVLM—fine-tuned with Low-Rank Adaptation (LoRA; $r=16$, $\alpha =32$, dropout = $0.05$) on the BreakHis dataset. Models were evaluated at $40\times$, $100\times$, $200\times$, and $400\times$ magnifications using accuracy, precision, recall, F1-score, and area under the ROC curve (AUC). While Qwen2 achieved moderate performance across magnifications (e.g., 0.8736 accuracy and 0.9552 AUC at $200\times$), SmolVLM consistently outperformed Qwen2 and substantially reduced the gap with CNN baselines, reaching up to 0.9453 accuracy and 0.9572 F1-score at $200\times$—approaching the performance of AlexNet (0.9543 accuracy) at the same magnification. CNN baselines, particularly ResNet34, remained the strongest models overall, achieving the highest performance across all magnifications (e.g., 0.9879 accuracy and 0.9984 AUC at $40\times$). These findings demonstrate that LoRA fine-tuned VLMs, despite requiring gradient accumulation and memory-efficient optimizers and operating with a significantly smaller number of trainable parameters, can achieve competitive performance relative to traditional CNNs. However, CNN-based architectures still provide the highest accuracy and robustness for histopathology classification. Our results highlight the potential of VLMs as parameter-efficient alternatives for digital pathology tasks, particularly in resource-constrained settings. Full article

This study comprehensively reviews the literature on knowledge management (KM) to explain its impact on organizational performance and business sustainability. It examines the dominant KM frameworks and theories; performance and sustainability outcomes; and key contextual enablers and constraints across sectors. Following the PRISMA 2020 guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analysis), a systematic review was used to find and collect relevant empirical and theoretical studies through Google Scholar, Scopus, and Web of Science. Thematic descriptive analysis of articles published between January 2020 and January 2026 revealed major themes, research trends, and conceptual gaps, which informed the key research agenda. A total of 70 studies were included after screening and eligibility assessment. The findings indicate that KM consistently enhances innovation capability and operational efficiency to boost competitive advantage and support social, economic, and environmental outcomes. These relationships are largely mediated by organizational learning and innovation, especially green innovation, and are moderated by leadership, organizational culture, and technological integration. Adoption patterns vary across industries and sectors based on differences in resources, digital maturity, and regulatory environments. Ongoing challenges include resistance to change, difficulties in managing tacit knowledge, measurement limitations, and limited longitudinal and cross-sectoral research. Overall, this systematic review highlights the need for integrated KM frameworks that align leadership, culture, and technology to strengthen performance and sustainability outcomes. It advances KM theory by clarifying the dominant models and mechanisms to offer actionable insights for managers and policymakers. Full article

The decarbonization of hard-to-abate sectors remains a significant challenge in achieving net-zero emissions targets. These industries depend on energy-dense fuels, making direct electrification and the direct use of hydrogen technically and economically challenging. Electrofuels present a promising pathway to reducing emissions while leveraging surplus renewable energy. This review evaluates the feasibility of electrofuels for deep decarbonization, focusing on production processes, energy demands, and economic viability. Environmental performance is discussed in terms of lifecycle greenhouse gas (GHG) emissions, carbon circularity considerations, and energy conversion efficiencies, while techno-economic feasibility is evaluated using metrics such as levelized cost of hydrogen (LCOH), CO₂ capture costs, and projected fuel production costs. The review indicates that while electrofuels can achieve substantial lifecycle emission reductions up to 40–90%, depending on pathway and electricity source, their deployment remains constrained by high energy demand, conversion losses, and capital costs. Projected reductions in LCOH to below $2.1/kg by 2030 and declining renewable electricity costs could significantly improve competitiveness, particularly in regions with abundant solar and wind resources. However, substantial trade-offs exist between efficiency, infrastructure compatibility, scalability, and carbon neutrality across different electrofuel routes. The review identifies key technological bottlenecks, cost drivers, and research priorities necessary to position electrofuels as a strategic solution for deep decarbonization in sectors where direct electrification is not feasible. Full article

The rapid growth of Arabic social media content requires the development of accurate and efficient methods for sentiment analysis. We propose a resource-efficient multi-task learning (MTL) framework for modern standard Arabic (MSA). The model uses a shared AraBERT encoder to jointly predict emotion, polarity, and intention. We integrate knowledge distillation (KD) from a large teacher model, self-distillation (SD) using model self-ensembling, and adversarial training (AT) as a regularization strategy. Experiments conducted on an annotated corpus of MSA tweets demonstrate that all distilled models outperform a fine-tuned multi-task baseline, and the combined KD+SD+AT configuration achieves competitive results. For instance, KD alone raised Macro F1 for emotion from 0.83 to 0.88 and for intention from 0.67 to 0.72. KD+SD+AT achieved the best intention F1 (0.76) and the highest polarity F1 (0.90). Notably, F1-scores for several minority classes show consistent improvement, particularly under KD and combined configurations. Paired t-tests confirm that several improvements, especially those obtained with KD and KD+SD+AT, are statistically significant ($p<0.05$). Our results indicate that distillation, combined with adversarial regularization, enables the development of smaller and more efficient Arabic sentiment models while maintaining competitive accuracy. These findings address a gap in Arabic multi-task sentiment analysis and provide a scalable, resource-efficient framework, along with empirical insights for distillation in Arabic language models. Full article

To address the challenges posed by multiple uncertainties in modern power systems to the market bidding of Virtual Power Plants (VPPs), this paper proposes an adaptive bidding strategy based on Deep Reinforcement Learning (DRL). First, a heterogeneous VPP aggregation model integrating dedicated energy storage, Vehicle-to-Grid (V2G), and flexible loads is constructed, incorporating complex physical and operational constraints. Second, to overcome the “myopic” local optimality problem of traditional DRL in temporal arbitrage tasks, a potential-based reward shaping mechanism linked to future price trends is designed to guide the agent toward long-term optimal strategies. Finally, multi-dimensional comparative experiments and mechanism analyses are conducted in a simulated day-ahead electricity market. Simulation results demonstrate the following: (1) The proposed algorithm exhibits robust convergence stability and effectively handles stochastic noise in market prices and renewable generation. (2) Economically, the strategy significantly outperforms the rule-based strategy and remains highly competitive with the deterministic-optimization benchmark under perfect-information assumptions. (3) Mechanism analysis further reveals that the DRL agent breaks through the rigid logic of fixed thresholds, learning a non-linear dynamic game mechanism based on “Price-SOC” states, thereby achieving full-depth utilization of energy storage resources. This work provides an interpretable data-driven paradigm for intelligent VPP decision-making in uncertain environments. Full article

This study evaluates the productive efficiency in the agrifood sector of 21 rural Bulgarian districts as a proxy for territorial competitiveness. Output-oriented Data Envelopment Analysis (DEA) was performed using district-level data from 2022 to 2024. The analysis incorporates five inputs related to labor, land, and capital and three economic outputs from agriculture and food processing. Results indicate substantial variation in efficiency among rural districts. Twelve districts form the efficiency frontier, with effective resource use and diverse structures; nine are inefficient due to scale or organizational/technological constraints. Bootstrap bias correction revealed standard DEA underestimates efficiency gaps. Frontier districts include large plains, mountainous regions and smaller, specialized systems, indicating diverse paths to competitiveness. A composite Territorial Competitiveness Index (TCI) showed frontier status does not guarantee efficiency, often due to underused manufacturing capital. Cluster analysis identified four performance groups needing different policy support, ranging from near-frontier territories that need knowledge transfer to deeply underperforming districts that require restructuring. No geographic clustering of efficiency was found, pointing to structural and institutional, rather than geographic, drivers. These results highlight the need for territorially tailored rural policies within the Common Agricultural Policy (CAP) and offer an empirical basis for diagnosing regional agrifood efficiency gaps. Full article

Optimizing nitrogen (N) management is crucial for high-quality rice (Oryza sativa L.) production. However, how N affects grain quality at different positions within a panicle remains unclear. This study evaluated the effects of different N application regimes on the milling, appearance, eating, and nutritional quality of grains at varying panicle positions. We used a japonica cultivar Wuyunjing 31 in a controlled pot experiment with three N treatments: N32:0 (early heavy N), N16:16 (split application with late N topdressing), and N16:0 (low-N control). Results showed that late N topdressing (N16:16) significantly improved head rice yield across all grain positions, which was linked to higher storage protein accumulation (especially glutelin) and larger length-to-width ratio. Conversely, late N application deteriorated appearance quality by increasing the chalky grain rate and chalkiness. This negative effect was most pronounced in superior grains on upper and middle branches. Furthermore, the N16:16 treatment consistently decreased amylose content while increasing albumin, prolamin, and glutelin levels, demonstrating a clear trade-off between carbon (C) and N sinks. We speculated that these intra-panicle differences result from increased competition for carbon resources between starch and protein synthesis pathways. Overall, precision N management should account for spatial differences in grain development to effectively balance rice yield and quality. Full article

Background: Small- and medium-sized enterprises (SMEs) constitute a large portion of the industrial energy demand in the emerging economies, but their shift to renewable energy is not well comprehended at the firm level. Bangladesh is a special case, since the country has adopted national commitments to Sustainable Development Goal 7 on clean energy, but the uptake of renewable energy by SMEs remains minimal due to complex socio-economic factors. Most of the literature has concentrated on household access to energy or national policy models, leaving a gap in empirically validated models of firm-level adoption in the manufacturing sector. Method: Based on the diffusion of innovation theory, institutional theory, and the resource-based view, this research paper formulates and empirically verifies a combined socio-economic model of renewable energy adoption. Partial least squares structural equation modeling (PLS-SEM) was used to analyze a cross-sectional survey of 426 owners and managers of manufacturing SMEs in Bangladesh’s textile and food processing sub-sectors. Findings: Four out of five hypothesized direct relationships were supported. The most important drivers were environmental orientation (β = 0.467, p < 0.001, f² = 0.413), market competitiveness (β = 0.287, p < 0.001, f² = 0.413), policy and institutional factors (β = 0.211, p < 0.001, f² = 0.413), and access to finance (β = 0.096, p = 0.004). Perceptions of cost did not become significant (β= −0.036, p = 0.279). Top management support significantly and negatively moderated the relationship between environmental orientation and adoption (β = −0.093, p = 0.003), possibly because it moderates the substitution mechanism in SME decision-making, which is highly centralized. The model accounted for 64.5% of the variation in renewable energy adoption (R² = 0.645). Conclusion: The results show that attitudinal and institutional factors tend to be more important than financial barriers in determining SMEs’ energy transitions. Environmental consciousness, market incentives, and streamlined institutional access should be the focus of policy interventions to hasten inclusive low-carbon transitions in emerging manufacturing economies. Full article