Search Results (21,913)

This study develops and empirically grounds the Human-Centered Transformation Framework (HCTF), an integrative model explaining how talent management (TM) functions as a dynamic capability aligning digital transformation (DT) and sustainability (SUS) within traditional manufacturing small- and medium-sized enterprises (SMEs) in the European Union. Integrating the Resource-Based View, dynamic capabilities theory, and Organizational Culture Theory, the framework was constructed through structured theory-building and validated using a mixed-methods sequential explanatory design. Quantitative data from 203 manufacturing SMEs across Poland, the Czech Republic, and Slovakia (78-item survey; Cronbach’s α = 0.84–0.91 across six constructs) provide statistical support for the framework’s core propositions, while qualitative interviews with 18 senior executives offer explanatory depth on the mechanisms through which TM enables transformation integration. Findings indicate that TM practice intensity is positively associated with both digital readiness (β = 0.42; p < 0.001) and sustainability maturity (β = 0.36; p < 0.001), with transformational leadership and learning-oriented organizational culture operating as significant mediating and moderating variables respectively. The study contributes a context-specific theoretical synthesis extending prior integrative TM models to the twin transitions context, while acknowledging limitations including the cross-sectional design and Central European sample. Full article

This study evaluates the application of deep learning techniques to the design of a microwave transmitter–receiver system operating in the 5G mid-band. The proposed architecture consists of four stages—signal generation, amplification, mixing, and filtering—each initially designed using conventional microwave methods and subsequently integrated into a complete transceiver. Simulation data were generated and component-specific convolutional neural networks (CNNs) were implemented in Python using TensorFlow/Keras. Across all models, an average error reduction exceeding 90% was achieved, with most networks converging after the third training cycle. System-level integration shows that the baseline design achieved a transmitted power of −32.637 dBm and a gain of 1.116 dB, while the deep learning-based design yielded −33.912 dBm and 0.738 dB. Additional analysis of S-parameters confirms acceptable impedance matching and a frequency response of around 3.5 GHz. These results illustrate that deep learning provides an effective complementary methodology for multi-component microwave system modeling and optimization in 5G applications. Full article

The increasing exploitation of sea cucumbers has driven widespread population declines, highlighting the need to improve knowledge and understanding of the early life history stages of exploited species such as Holothuria tubulosa, one of the most common holothurians along Mediterranean coasts. This study investigated larval settlement success and juvenile early survival of H. tubulosa larvae, considering two algal biofilms as settlement cues: the diatom Amphora sp. and the green alga Ulvella lens. Larvae were reared under controlled hatchery conditions, and, once reaching the doliolaria stage, larvae were individually exposed to biofilm-conditioned substrates vs. a control without biofilm. Settlement dynamics and larval development were monitored over 35 days and analysed using generalised linear mixed models, while the biochemical composition of the biofilms was assessed through protein, carbohydrate, and lipid quantifications. Larvae exposed to algal biofilms successfully settled and metamorphosed, whereas no settlement occurred in the control. U. lens induced the highest settlement success (54%) and supported subsequent juvenile development, while Amphora sp. resulted in lower settlement rates (21%) and higher post-settlement mortality. Although Amphora sp. showed higher protein and carbohydrate content, settlement and survival were enhanced on U. lens, suggesting that biofilm structure and biochemical cues play a primary role in regulating settlement processes. These findings improve the understanding of settlement mechanisms in H. tubulosa and provide valuable insights for hatchery production, conservation strategies, and the sustainable aquaculture of Mediterranean sea cucumbers. Full article

The hematite phase decorated with iron-doped cerium oxide nanoparticles (F@FC) was precipitated from cerium and iron oxalate intermediate products. The photocatalytic composite of graphitic carbon nitride (gCN) and F@FC was prepared by a simple method involving mixing the two components, followed by thermal treatment at 400 °C. According to electron microscopy, F@FC is composed of a submicron iron oxide (hematite) phase decorated with iron-doped cerium oxide nanoparticles deposited on gCN substrate. A hierarchically structured composite was observed instead of a simple mechanical mixture of α-Fe₂O₃, Fe-CeO₂, and gCN. To observe two types of degradation activity, photocatalytic and Photo-Fenton degradation activity, Rhodamine B (RhB) was applied as the model water pollutant. The influence of the amount of photocatalyst, the RhB concentration, the presence of cations and anions, the pH, and the effect of e⁻, h⁺, •OH, and •O₂⁻ scavenging reactants were studied. The Photo-Fenton degradation exhibited high efficiency across the entire tested pH range, whereas photocatalytic degradation showed comparable activity only at acidic pH. The F@FC-gCN composite catalyst exhibited a high degree of recyclability. The degradation pathways of photocatalytic and Photo-Fenton reactions were suggested by HPLC-MS analysis of the reaction products. A notable finding of this study was the observation that the green-yellow, fluorescent intermediate Rhodamine 110 was formed during the photocatalytic degradation of RhB. However, the high reactivity of the generated •OH radicals during Photo-Fenton degradation has been demonstrated to inhibit the formation of intermediate Rhodamine 110. Full article

Industrial defect detection is a fundamental task in intelligent manufacturing, yet existing object detection methods often struggle with the characteristics of industrial defects, such as small size, irregular shapes, and complex visual backgrounds. Moreover, most detection models are designed primarily for natural image datasets, resulting in limited robustness when deployed in real-world industrial environments. To address these challenges, this research focuses on industrial defect detection and presents contributions at both the dataset and method levels. First, two real-world industrial defect datasets collected from actual production lines are introduced, namely, the Stator Housing Defect Dataset and the Cover Plate Silicone Defect Dataset, which cover representative inspection scenarios with distinct defect characteristics. Second, we propose a detection transformer with a mixed encoder for industrial scenarios (IM-DETR). By integrating heterogeneous multi-scale feature representations, the proposed framework jointly enhances local detail sensitivity and global contextual reasoning without relying on complex post-processing. Extensive experiments on the proposed industrial datasets demonstrate that IM-DETR consistently outperforms existing state-of-the-art detection methods, particularly in scenarios involving small defects, complex backgrounds, and appearance ambiguity, validating the effectiveness and robustness of the proposed approach. Full article

This study examines the compressive behavior and analytical modelling of natural and rubberized concretes (RuC) confined with low-cost glass chopped-strand mat (GCSM) jackets. A total of forty-two cylindrical specimens were tested under axial compression to assess the influence of rubber particle size, confinement configuration, and the number of GCSM layers. The RuC mixes were prepared by replacing 20% of fine aggregate by volume with crumb rubber of two size fractions: coarse (2.0 mm, retained on #10 sieve) and fine (0.425 mm, retained on #40 sieve). Both full- and strip-wrapping schemes were applied using two, four, and six layers of GCSM. The results demonstrated that GCSM jackets significantly enhanced the mechanical performance of both NAC and RuC specimens. Full wrapping provided the highest confinement efficiency, increasing compressive strength by up to 115% for NAC and 90% for RuC, while the ultimate axial strain increased by more than 1300% compared with unconfined specimens. Strip wrapping also improved performance, producing strength gains of 25–45% and strain increases of 250–500%. Analytical stress–strain models were developed through regression analysis, showing strong correlation with the experimental results (R² = 0.80–0.99). The proposed GCSM jacket system demonstrates high potential as a sustainable and economical alternative for strengthening and retrofitting rubberized concretes, offering improved ductility and energy absorption while supporting circular material utilization. It is noted that the confinement ratio, size of rubberized aggregates, and their percentage replacement of rubberized aggregates should be consistent with the values used in this work in order to use the proposed analytical expressions. Full article

(1) Background: Colombia hosts one of the world’s largest mixed-displacement crises, combining longstanding internal displacement with the influx of Venezuelan migrants. This case study examines how the Emergency Medical Team (EMT) Hospital Barco San Raffaele (HBSR) adapted its service-delivery model to respond simultaneously to internal displacement in the Colombian Pacific region and the Venezuelan refugee influx. Using the WHO EMT Surge Capacity Framework, the study analyses how health services were adapted across two concurrent displacement contexts. (2) Methods: A mixed-methods comparative case study was conducted using mission reports, epidemiological surveillance data, policy reports and institutional documents collected between November 2020 and May 2021. Data were analyzed through a thematic analysis structured around the four domains of the WHO EMT Surge Capacity Framework (Staff, Structure, Supplies and Systems), to examine how service adaptation was operationalized across different geographic, sociocultural and legal environments; (3) Results: EMT HBSR adapted staffing composition, supply chains, infrastructure, and operational systems across both settings. Its hybrid model, combining a hospital boat platform with mobile outreach teams, enabled continuity of primary care, mental, maternal and child health, and community-based services in geographically isolated and culturally diverse communities; (4) Conclusions: The findings illustrate how flexible EMT operational models can support the adaptation of health services, and reduce health access inequalities in displacement contexts characterized by high mobility, confinement and limited health system capacity. Mobile platforms, such as hospital boats, appear to be a viable strategy for ensuring continuity of care along migratory routes and in geographically isolated areas affected by protracted instability. Full article

Remaining useful life (RUL) prediction is essential for ensuring equipment reliability in smart manufacturing. However, mixed-model production introduces a significant challenge due to the discrepancy between the continuous nature of latent degradation and the abrupt, discrete transitions observed in sensor signals. These transitions are driven by the stochastic sequencing of product variants, which obscures the true health state of the equipment. Traditional RUL models are primarily designed for continuous and coherent evolutionary patterns, and consequently, they struggle to distinguish these observable, event-driven jumps from the hidden, underlying degradation trajectories. To resolve this, we propose the Dynamic Segmentation Network (DSNet), a framework designed to synchronize with discrete production rhythms while preserving the continuity of latent health indicators. Specifically, a segmentation loss integrating Proxy-NCA and information entropy is developed to guide the model in discerning discrete process boundaries and achieving semantically consistent partitioning. Furthermore, a hybrid encoding scheme integrates absolute and rotary positional information to capture multi-granularity temporal dependencies, which effectively bridges global degradation trends with local intra-segment structures. These innovations empower DSNet to extract highly discriminative features that are robust to process-induced fluctuations, thereby significantly enhancing RUL prediction performance. Extensive evaluations on 53 industrial welding guns from Bayerische Motoren Werke (BMW) Shenyang plants demonstrate that DSNet achieves reductions in MAE and RMSE by 12.29% and 10.66%, respectively. Consistent performance gains across three public benchmarks further validate the framework’s exceptional generalizability and robustness. Full article

With the rapid growth of the civil aviation industry, airport gate resources—especially those equipped with jet bridges (more convenient than shuttles)—have become increasingly scarce, posing new challenges to the sustainable management of airport operations. In a real-world application of the airport transport optimization study field, the airport gate assignment problem (AGAP) has emerged as a critical scheduling task in airport operations with the rapid growth of passenger demand. In this study, a mixed-integer linear programming model is developed for AGAP, aiming to minimize baggage transfer vehicle usage, maximize airline satisfaction, reduce passenger boarding time, and enhance the overall sustainability of airport operations. To efficiently address the computational complexity of this integrated modeling framework, a customized multi-objective particle swarm optimization (MOPSO) algorithm is proposed, augmented by a tabu search (TS) strategy. The TS algorithm provides high-quality initial solutions for MOPSO and performs local intensification on elite particles, thereby enhancing both convergence speed and solution quality. Extensive numerical experiments demonstrate that the proposed hybrid approach significantly outperforms the standalone MOPSO algorithm, achieving a 26.37% improvement over the original gate assignment scheme and a further 1.25% improvement compared to the standalone MOPSO, confirming the effectiveness and practicality of the proposed method. Full article

Tree ensembles—Random Forests (RFs) and Gradient Boosting Machines (GBMs)—often stabilize before all trees are evaluated. We study early stopping as a nonparametric change-point problem on prediction increments. The P²-STOP method family monitors a robust interquartile-range (IQR) scale of prediction increments online and stops when a relative-scale criterion is met. The default variant uses a rolling-window exact-quantile estimator ($O\left(w\right)$ memory), which provides a clean finite-sample stopping guarantee; a full-prefix P² streaming approximation ($O\left(1\right)$ memory) is available as a memory-light alternative. The stopping rule applies to both RFs and GBMs without model-specific distributional assumptions. On four RF benchmarks (MNIST, Covertype, HIGGS, and Credit Card Fraud), P²-STOP achieves 44.8% mean work reduction (range: 0.7–71.7%) with an accuracy change from $-0.53$ to $+0.02$ percentage points versus full-ensemble inference. On XGBoost ($T=500$), work reduction is dataset-dependent (41.4% on Covertype up to 89.0% on Credit Card), with corresponding accuracy trade-offs. Under random-tree contamination conditions (5%, 15%, and 25%), performance remains stable, whereas IQR-versus-standard-deviation baseline differences are mixed rather than uniformly dominant. Designed for compiled inference engines (e.g., C++/Numba), P²-STOP translates theoretical work reduction into consistent wall-clock speedups ($4.14\times$–$4.82\times$ versus compiled full RF on MNIST/Covertype/HIGGS for $T=500$). Native Python implementations serve purely as logical baselines due to loop overhead, while Credit Card exhibits the expected slowdown when work reduction is near zero. All comparisons use five seeds with 95% confidence intervals and seed-level paired tests. With only five seeds, inferential power is limited, and p-values should be interpreted cautiously. Relative to the Dirichlet RF baseline, our contribution is not larger RF-specific work reduction; it is a robust nonparametric IQR-scale stopping criterion, cast as a change-point/sequential-inference problem, that works as a post hoc wrapper across RF and GBM settings. Full article

Cancer center infusion departments are often challenged with scheduling a large number of patients while having a limited number of nurses available to administer the infusions. Cancer patients have different acuity levels depending on many factors, such as treatment plans, drug side effects, and health status. Thus, several factors need to be considered when assigning patients to nurses, as unbalanced nurse-to-patient assignments affect patient flow and nurse workload. This study introduces a mixed-integer programming model for nurse–patient assignments that minimizes patient wait times while ensuring workload balance among oncology nurses, while addressing the limited attention in existing studies to jointly modeling patient acuity and nurse continuity. The model also explores the effects of maintaining nurse continuity for patients desiring the same nurse throughout their treatments. Because the mixed-integer programming model can become difficult to solve when there are many cancer patients, an alternative nurse–patient assignment heuristic is proposed and evaluated. Numerical examples based on data from a regional cancer center compare the effectiveness and performance of the exact and heuristic methods. The results show that patient wait time and workload variation among nurses increase when there is a stronger requirement to maintain nurse continuity, which could negatively affect both patient and nurse satisfaction. This study provides valuable insights into the nurse–patient assignment problem and helps cancer infusion centers determine the impacts of maintaining different levels of nurse continuity in their settings. Full article

Given the therapeutic potential of bioactive cordycepin in medical and healthcare products, precision fermentation using an engineered strain of Aspergillus oryzae was performed to enhance cordycepin production. To understand and predict the dynamics of cell growth and cordycepin production in this fungal strain, mathematical modeling of submerged fermentation was applied. The effects of different nitrogen sources (yeast extract, peptone, (NH₄)₂SO₄, NH₄Cl, NaNO₃, and KNO₃) and carbon sources (glucose and cassava starch hydrolysate, CSH) on cell growth and cordycepin production were evaluated under submerged fermentation conditions. The results showed that organic nitrogen sources significantly enhanced biomass formation and cordycepin production compared with inorganic nitrogen sources. Among them, yeast extract provided the best performance, yielding the highest biomass (13.63–15.99 g/L) and cordycepin titer (1.24–1.72 g/L). In contrast, nitrate-based nitrogen sources supported cell growth but resulted in negligible cordycepin production. Under optimized conditions in a bioreactor, both glucose and CSH supported fungal growth, although CSH promoted higher biomass formation while glucose favored cordycepin biosynthesis. The kinetic model demonstrated that the growth of engineered A. oryzae was well described by the logistic growth model (R² > 0.88). The cordycepin production profiles were well fitted by the Luedeking–Piret model (R² > 0.99), indicating a mixed growth-associated product with kinetic constants α and β representing growth-associated and non-growth-associated production, respectively. Overall, the developed kinetic model provides a quantitative framework for describing cell growth, substrate utilization, and cordycepin formation, offering guidance for process optimization and scale-up of cordycepin production in engineered fungal systems. Full article

This study investigates the preparation and performance of ultra-high-performance concrete (UHPC) incorporating manufactured sand as a full replacement for quartz sand. The mix design was optimized by integrating the compressible packing model (CPM) with an orthogonal experimental design. The influence of stone powder content in manufactured sand—0, 5, 10, and 15% by mass of fine aggregate—on fresh-state fluidity and 7d-mechanical properties was systematically evaluated. Hydration products and microstructural features were analyzed using X-ray diffraction (XRD), scanning electron microscope (SEM), and mercury intrusion porosimetry (MIP). Results show that the manufactured sand-based UHPC achieved a fresh-state fluidity of 185 mm and a 7-day compressive strength of 152.4 MPa. Both fluidity and compressive strength exhibited a unimodal trend with increasing stone powder content, reaching maxima at 10%. Microstructural analysis revealed intimate interfacial bonding between unhydrated particles and calcium silicate hydrate (C–S–H) gel; notably, the UHPC matrix with 10% stone powder displayed the densest microstructure. MIP results further demonstrated that an optimal stone powder content effectively reduced total porosity, with the lowest overall porosity and the highest volume fractions of harmless (≤20 nm) and less harmful (20–100 nm) pores observed at 10%. These microstructural refinements collectively underpin the superior mechanical performance of manufactured sand-based UHPC. Full article

With the increasing use of reinforced concrete segments in large-scale tunnels, engineering projects have placed higher mechanical demands on concrete, and the choice of concrete materials significantly influences these mechanical properties. This study is based on the preliminary mix design for the concrete used in the Second Undersea Tunnel Project, with the mass content of nano-SiO₂ (NS) (1–3%), the volume content of steel fibers (SF) (0.5–1.5%) and the volume content of polypropylene fibers (PPF) (0.05–0.25%) as independent variables and using compressive strength (Y₁), splitting tensile strength (Y₂), and toughness index (Y₃) as response variables. Using the Box–Behnken response surface design method, response surface models for each parameter were established and analyzed. The effects of NS, SF, and PPF on the mechanical properties of the concrete were investigated. Combining the MOPSO algorithm and the entropy-weighted TOPSIS method, a multi-objective cooperative optimization study was conducted. Finally, a microstructural analysis of the optimal NSDHFRC was performed. The results indicate that Y₁, Y₂, and Y₃ all initially increase and then decrease with increasing NS content; Y₁ and Y₃ increase with increasing SF content. However, when the SF content exceeds a certain level, the fiber spacing becomes too dense, weakening the effective bridging effect between fibers, resulting in a decrease in Y₂ at excessively high SF contents; PPF can suppress crack formation within a certain content range, but its effect on Y₁ is relatively weak. Due to agglomeration and water absorption, both Y₂ and Y₃ decrease when the PPF content is too high. It was determined that the optimal solution occurs when the mass fraction of NS is 2.15%, and the volume fractions of SF and PPF are 1.37% and 0.063%, respectively, with Y₁, Y₂, and Y₃ being 69.94 MPa, 5.49 MPa, and 1.99, respectively. Experimental verification confirmed that the relative error is within 5%. A microscopic analysis of the optimal solution revealed that an appropriate amount of NS refines the concrete structure through physical and chemical reactions, improves the interface transition zone, and enhances the bond strength between the fibers and the matrix. Meanwhile, PPF and SF distribute stress, respectively delaying the propagation of microcracks and macrocracks during different loading stages. These findings provide a reference for practical engineering applications. Full article

Lung cancer is one of the most common causes of death from cancer around the world, and catching it early through computed tomography (CT) scans can drastically improve survival. However, automated classification of pulmonary nodule candidates is hard because images do not all have the same intensity across scanners and protocols, resulting in inconsistent performance, more false positives (FP), and a ceiling on how much deep learning models work in an average clinic. In this work, we tackle this by introducing a preprocessing step that corrects intensity differences before feeding images into classification models. We use Contrast-Limited Adaptive Histogram Equalization (CLAHE), but with its key parameters tuned automatically via a modified version of the Covariance Matrix Adaptation Evolution Strategy (CMA-ES). This helps to boost local contrast adaptively, keeps important anatomical details intact, and cuts down on noise. We tested the approach on the public LUNA16 dataset, first checking image quality (Peak Signal-to-Noise Ratio (PSNR) around 53 dB and Structural Similarity Index (SSIM) of 0.9, better than standard methods), then training three popular deep models—namely, ResNet-50, EfficientNet-B0, and InceptionV3—with CutMix augmentation for better generalization. On the enhanced images, ResNet-50 achieved up to 99.0% classification accuracy with substantially less FP than when using the raw scans. Taken together, these results demonstrate that intelligent and optimized preprocessing can effectively mitigate intensity variations via deep learning for lung nodule detection, thus coming closer to realizing the practical toolbox of computer-aided diagnosis in routine clinical practice. Full article