Search Results (12,105)

Although Industry 4.0 has successfully advanced lean manufacturing through digitalization and automation, its primary focus on operational efficiency leaves emerging strategic priorities—human-centricity, sustainability, and resilience—outside its original scope. The Industry 5.0 agenda explicitly elevates these three pillars, creating new potential to drive lean transformation. However, how Industry 5.0 can systematically drive lean manufacturing transformation remains unclear. To address this knowledge gap, this study develops a strategic roadmap. First, a content-centric literature review identifies 12 key enablers for Industry 5.0-driven lean manufacturing. Second, Fuzzy Interpretive Structural Modeling (FISM) and expert opinions determine hierarchical relationships among the enablers and construct a multi-level structural model. Third, Matrices d’Impacts Croisés Multiplication Appliquée à un Classement (MICMAC) analysis evaluates the driving power and dependence of each enabler. Finally, a strategic roadmap is developed based on expert synthesis. The findings reveal that “government regulation and incentives” and “employee skill training” are the most critical enablers, while “value chain design and improvement” and “resource input and return” are the most complex and difficult to develop. The roadmap highlights the mediating role of “stakeholder participation and collaboration.” Importantly, the roadmap addresses potential tensions in lean implementation—such as the carbon footprint trade-off of frequent small-batch transport—by embedding sustainability assessment into value chain design and technology governance. This study offers a practical guide for manufacturers to prioritize investments and sequence actions toward lean transformation in the Industry 5.0 era. The main contribution of this study is a strategic roadmap that explains how Industry 5.0 can enable lean manufacturing transformation through prioritized actions and hierarchical enablers, while reconciling efficiency with sustainability and resilience goals. This roadmap offers a practical guide for manufacturers and policymakers to sequence investments and actions toward lean transformation in the Industry 5.0 era. Full article

The positioning and clamping process in aircraft assembly exhibits pronounced long-term temporal correlations and intense human–machine interactions. Consequently, assembly quality depends heavily on operator compliance and consistency. Capturing long-term, strongly correlated features in complex industrial environments remains a significant challenge. To overcome this, this study proposes a Long-Term Strongly Associated Action Recognition Network (LTSA-Net) tailored for aircraft assembly positioning and clamping tasks. Based on the C3D backbone, the model first incorporates the SimAM attention mechanism and BN modules to significantly enhance focus on critical spatiotemporal features. To address the challenge of capturing long-term temporal dependencies, LTSFEM is designed to extract global temporal information accurately. Furthermore, to balance structural lightweight design with real-time inference requirements, the CWSTB module is integrated to achieve substantial parameter compression. In addition, a dedicated aircraft assembly positioning and clamping dataset was constructed, and a robust training framework was established using the AdamW optimizer and Mixup data augmentation. Experimental results demonstrate that LTSA-Net achieves a recognition accuracy of 98.82% on the LTSA-Dataset, with a per-frame inference time of 42 ms, successfully meeting the dual requirements of high precision and real-time performance in industrial scenarios, and providing a practical technical solution for intelligent monitoring of aircraft assembly processes. Full article

Current landslide displacement prediction models often suffer from insufficient integration between physical mechanisms and data-driven approaches, weak model generalizability, and limited operational applicability. To address these issues, this study develops a physics-informed online learning framework for landslide displacement prediction. The core of this framework is a Physics-informed Long Short-Term Memory network (Phys-LSTM). By embedding discretized forms of the stress balance, creep constitutive, and kinematic equations as hard constraints into the LSTM’s gating mechanisms and loss function, the model ensures physically consistent predictions and enhanced interpretability throughout the learning process. Leveraging real-time data streams from the Sichuan Provincial Geological Hazard Monitoring and Warning Platform, we developed an online processing pipeline for real-time multi-source data ingestion, automated quality control, spatiotemporal alignment, and physics-informed feature engineering. A progressive three-stage learning algorithm was designed to support model cold-start, incremental training, and rolling prediction. Validation across 45 model-development landslide sites and one independent application case demonstrated the framework’s significant superiority over traditional models in displacement prediction accuracy (RMSE ≤ 1.78 mm, R² ≥ 0.96), cross-site generalization stability, and its capability to capture accelerated deformation phases. This research indicates that deeply integrating geomechanical prior knowledge into an online learning framework can effectively improve the reliability, interpretability, and operational applicability of landslide displacement prediction models, thereby providing methodological support for subsequent landslide early warning applications. Full article

The economic resilience of agricultural enterprises is increasingly relevant for maintaining processing continuity and food quality in highly perishable agro-food chains. This study examines the associations between financial knowledge, financial management competency, business liquidity, and operational food-processing continuity in Türkiye’s tea sector. A quantitative cross-sectional design was employed, using structured survey data from 203 senior managers across 86 public and private tea-processing firms in Rize Province. The data were analysed using Ordinary Least Squares regression, mediation analysis, exploratory factor analysis, and robustness checks in accordance with OECD/INFE guidelines. Results indicate a significant deficit in theoretical financial knowledge (mean score: 4.47/10) alongside widespread overconfidence among 85% of managers. Applied financial management competency is positively associated with perceived business liquidity (β = 0.336, p < 0.001), suggesting that practical budgeting, cash-flow planning, and financial decision-making capabilities are relevant to maintaining operational funding capacity. In contrast, cash-flow difficulties are not significantly explained by firm-level financial knowledge, managerial competency, liquidity, or ownership structure (R² = 0.014, p = 0.722), indicating that these difficulties may reflect broader seasonal and sector-wide financing constraints. The findings challenge the assumption of a linear relationship between theoretical financial knowledge and managerial outcomes. They suggest a dual policy approach that combines applied financial management training with structural financing mechanisms to ensure the continuity of fresh leaf procurement and processing. While the study does not directly measure food safety, post-harvest losses, or SDG outcomes, the results have potential implications for reducing processing disruptions and supporting more resilient agro-food processing systems. Full article

Soluble solids content (SSC) is one of the most important indicators of sweetness, ripeness, and market quality in sweet cherries. However, conventional SSC determination is destructive, labor-intensive, and unsuitable for rapid or large-scale quality assessment. Therefore, there is a need for fast, non-destructive, and data-driven sensing approaches that can estimate internal fruit quality without damaging the sample. This study aimed to develop a non-destructive approach for SSC prediction in sweet cherries by combining open-ended coaxial probe dielectric spectroscopy with deep learning models. An open-ended coaxial probe measurement system was designed and developed to determine the dielectric properties of sweet cherries and was coupled with an Agilent E4991A impedance analyzer operating over a frequency range of 5–3005 MHz. A total of 10,080 dielectric measurements and 2100 reference SSC measurements were collected over 26 experimental days. The dielectric constant (ε′), loss factor (ε″), and loss tangent (tan δ) were extracted and used to construct separate ε′, ε″, tan δ, and integrated combined datasets. Six deep learning architectures, namely convolutional neural network (CNN), long short-term memory (LSTM), bidirectional long short-term memory (BiLSTM), gated recurrent unit (GRU), CNN-LSTM, and convolutional long short-term memory (ConvLSTM), were trained and optimized using Bayesian optimization and early stopping. CNN achieved the best performance on the tan δ dataset (test R² = 0.9099, RMSE = 0.8354 °Brix, MAE = 0.6599 °Brix), whereas GRU yielded the highest accuracy on the integrated combined dataset (test R² = 0.8622, RMSE = 1.0331 °Brix, MAE = 0.7958 °Brix). ConvLSTM provided the most consistent performance across all four datasets (test R² = 0.8081–0.8651), demonstrating strong predictive capability and practical computational efficiency. These findings confirm the potential of reduced-range dielectric spectroscopy combined with deep learning for rapid, non-destructive SSC assessment in sweet cherries. Full article

Teacher–AI collaboration is increasingly present in educational settings, yet little is known about how it is conceptualized in empirical research and what this implies for teacher preparation. This review synthesizes 39 empirical studies on teacher professionalization published between 2015 and 2025 to examine how responsibilities for detecting, diagnosing, and acting on educational data are distributed between teachers and AI systems. Results indicate a predominant focus on learning analytics and natural language processing tools, largely operating at intermediate to high levels of automation. In these configurations, teachers are primarily positioned as interpreters or monitors of AI outputs. In addition, our analysis identifies a consistent pattern: as AI systems assume greater pedagogical autonomy, teacher training described in the literature remains brief, procedural, and largely limited to technical familiarization. These findings suggest that different automation configurations entail distinct competence demands, and that teacher preparation must move beyond technical training to conceptualize teacher–AI collaboration as ongoing professional sensemaking within hybrid intelligent systems grounded in educational values. Full article

Urban low-altitude unmanned aerial vehicle (UAV) logistics networks face critical operational bottlenecks due to complex three-dimensional spatial blockages, continuous communication diffraction, and severe vulnerability to information-layer threats such as Automatic Dependent Surveillance—Broadcast (ADS-B) signal anomalies. To address these interconnected challenges, this paper proposes an event-driven, cloud–edge collaborative digital twin framework to guarantee continuous multi-link communication and flight safety. The architecture operates through a dual-tier “Teacher–Student” paradigm. Under secure conditions, a cloud digital twin acts as a high-capacity “Teacher,” employing Density-Based Spatial Clustering of Applications with Noise (DBSCAN) to partition heterogeneous user topologies. It then utilizes an energy-guided stochastic diffusion sampling (EGSDS) method to refine initial macroscopic routing, generating precise, outage-free global trajectories by systematically minimizing non-line-of-sight (NLoS) observation penalties and kinematic regularization costs. To counteract signal anomalies, a distributed Time Difference of Arrival (TDOA) anchor network continuously validates UAV coordinate integrity. If a threshold is breached, control authority is instantly transferred to the UAV’s edge digital twin. This resource-constrained edge tier relies on a localized “Student” network trained via progressive distillation. By compressing the computationally heavy iterative diffusion process into a rapid one-step inference model, the UAV autonomously generates a secure, short-range emergency path that strictly adheres to minimum communication thresholds. Once interference clears, the cloud seamlessly regains control to complete the logistics mission. Experimental results demonstrate that the proposed scheme significantly outperforms conventional heuristic routing methods in cloud-based scenarios. Furthermore, the edge-based distillation mechanism substantially improves the overall trajectory survival rate under signal anomalies, ensuring resilient and continuous logistics operations. Full article

A virtual training environment offers clear advantages for agricultural robotics. It provides a safe setting in which perception, navigation, and control algorithms can be evaluated without risking damage to either the robot or the crop. It also supports efficient data generation: large volumes of training data can be collected under diverse environmental conditions that would be costly, slow, and often season-dependent in real-world deployments. This broader variability improves model adaptability, reduces the risk of overfitting, and leads to more robust operation. In this paper, we argue that digital twin technology should therefore be understood not merely as a passive mirror of a physical robot, but as an active training environment in which multiple sensor-related subprocesses can be developed, tested, validated, and refined jointly. This paper is based on our experiences with digital twin technology used in the development of a vineyard robot, including a self-driving rover, sensor simulation, procedural map generation, and agriculture-specific movement models. Our contribution is threefold: we reinterpret the digital twin as a training space, propose a layered framework for training agricultural robots in virtual environments, and explain why agriculture is a particularly strong use case, given variable field conditions, expensive real-world experimentation, and persistent labor scarcity. To validate this framework, we present the simulation-based evaluation of an autonomous reinforcement learning agent. The agent has been trained entirely in this virtual environment, which successfully navigated to 155 out of 161 target points in a simulated vineyard demonstration environment. Full article

Disability Enterprises have the potential to address employment barriers that people with an intellectual disability experience as they move into open employment. Disability Enterprises are able to facilitate this transition through strategic organisational design, but it is unknown the extent to which organisations are following what could be considered best practice. Utilising a survey and interview approach, we aimed to understand how well organisations align with the ‘WISE-Ability’ model and the ongoing challenges they face in supporting open employment pathways for people with a disability working at the enterprise. Staff (94) from Disability Enterprises completed a survey where they rated their own organisation against a number of criteria related to organisational design and operation related to transitioning supported employees to open employment. After completing the survey, organisational staff (19) participated in a semi-structured interview. Disability Enterprises provide training and life skills development options with the end goal of employment transition. Training is adapted to the needs of individuals and there is flexibility in the pace of learning and rostering of shifts. Disability Enterprises develop industry-specific work skills and independent life skills. Pathways to employment are offered in most cases; however, there is variation in the success and scale of employment pathways. Organisations continue to face challenges that exist in the disability service system and open labour market. Organisations are confident that they are able to offer a culture of support and respect, choice and variety of employment options, busy and quiet spaces, and areas for rest and accessible workspaces where individuals feel empowered and safe to try new tasks. Many organisations developed relationships with external stakeholders and employers to facilitate financial sustainability and employment pathways. However, organisations face challenges in the following areas: resourcing pathways to employment and offering certified training options for people working in a Disability Enterprise. Employment pathways were often carved out on a case-by-case basis relying on significant staff support and after-hour work. Full article

Highway–rail grade crossing (HRGC) incidents remain a persistent safety concern due to repeated interactions between roadway users and rail operations under varying environmental and operational conditions. Existing studies rely on raw incident counts or partial exposure measures that can be influenced by differences in infrastructure exposure and do not account for spatial dependence, limiting consistent comparison across locations. This study developed an exposure-normalized framework to model incident intensity at the county level using accumulated incidents per crossing (AIPC), which normalizes cumulative incidents by crossing exposure. The analysis integrated statistical distribution modeling, spatial clustering, and supervised machine learning. The study combined county-level HRGC data for the contiguous United States from 1975 to 2025 with infrastructure, traffic, environmental, and accessibility variables. Results showed that AIPC was consistent with a gamma distribution, indicating a continuous representation of incident intensity without discrete risk regimes. Local Moran’s I identified statistically significant high-intensity clusters in specific regions, confirming spatial dependence in incident intensity. Machine learning models achieved strong predictive performance, with the extra trees model reaching AUC = 0.907 (F1 = 0.528) and ensemble methods consistently outperforming linear and kernel approaches. SHAP and permutation-based feature importance analysis identified temperature, train frequency, and accessibility measures as the most influential predictors, while aggregate density measures contributed the least. The results provided consistent evidence that incident intensity was associated with environmental conditions, operational exposure, and network structure. The proposed framework supports exposure-based risk assessment and enables identification of high-intensity counties for targeted intervention. This approach provides a transparent and transferable method for improving HRGC safety analysis and prioritizing resource allocation across large geographic areas. Full article

In transient stability preventive control of power systems, time-domain simulation is computationally intensive. In addition, the initial operating feature data often contain abundant redundant and irrelevant information. These factors may adversely affect the assessment performance of machine learning models. To address these issues, a transient stability preventive control method based on the sample convolution and interaction network (SCINet) is proposed. First, a feature selection algorithm based on the orthogonal maximal information coefficient and information gain (OMICIG) is developed to extract the key operating features of the system. Second, the SCINet model is employed to learn the nonlinear mapping relationship between the selected key operating features and the transient stability index (TSI). Then, the trained SCINet model is embedded into the transient stability constrained optimal power flow (TSCOPF) model as a surrogate transient stability constraint. In this way, the complicated computation associated with nonlinear differential-algebraic equations (DAE) in the conventional TSCOPF model is avoided. Furthermore, an improved dung beetle optimizer (IDBO) algorithm is used to iteratively solve the resulting model, thereby deriving a preventive control strategy that ensures transient stability while maintaining system operating economy. Finally, simulation studies on the New England 10-machine 39-bus and the IEEE 118-bus system demonstrate the effectiveness of the proposed method. Full article

Low-voltage electrical contacts are core components of power distribution systems, renewable energy installations, and industrial automation equipment. The electric arc generated during contact switching is the primary cause of contact erosion, material transfer, and equipment failure, posing significant threats to system reliability and operational safety. The accurate prediction of arc parameters is hindered by two challenges: the high scatter in available data undermines empirical models, and purely data-driven approaches risk physically implausible results. To address this, a Gaussian Mixture-enhanced Bayesian-optimized Physics-Informed Neural Network (GB-PINN) is proposed. Three core contributions are made: (1) High-fidelity MHD simulation foundation: A magnetohydrodynamic (MHD) multi-physics coupling model of the contact arc was constructed and validated against experiments, showing high fidelity with only 1.63% error in arc duration and 1.82% in arc energy. A multivariate simulation dataset was generated by varying key contact parameters based on this validated model. (2) GMM-based data augmentation: The measured and simulated data were modeled and sampled via Gaussian Mixture Model (GMM) to enrich the dataset while preserving physical consistency. (3) BOHB-optimized PINN prediction: The Bayesian Optimization and Hyperband (BOHB) algorithm was employed to optimize the PINN hyperparameters, enhancing training efficiency and predictive accuracy. Experimental results demonstrated that the proposed GB-PINN achieved superior performance in predicting arc duration and energy, with mean absolute errors (MAE) of 0.079 ms and 0.624 mJ, root mean square errors (RMSE) of 0.099 ms and 0.774 mJ, and coefficients of determination (R²) of 0.980 and 0.979, significantly outperforming grey model (GM (1, N)), long short-term memory (LSTM), and Transformer models. As a physics-informed data-driven tool, GB-PINN enables high-precision arc prediction, providing reliable support for electrical contact design. Full article

Background/Objectives: Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis (PEP) is a clinically significant complication of ERCP, occurring in approximately 2–10% of general cases and at higher rates in high-risk patients. Early prediction of PEP risk may support timely intervention and improved patient management. This retrospective single-center study comparatively evaluated tabular clinical data, endoscopic image data, and multimodal fusion approaches for PEP prediction. Methods: Retrospective data collected from the Sindh Institute of Advanced Endoscopy and Gastroenterology were analyzed using machine learning and deep learning techniques. XGBoost(version 3.2.0) was applied to tabular clinical data, while EfficientNet-B0, ResNet50, and DenseNet201 were used for endoscopic image analysis. A multimodal contrastive learning (MMCL)-based framework combining ResNet50 image features with multilayer perceptron (MLP)-based tabular features was additionally implemented for binary PEP prediction. Class imbalance mitigation techniques, including data augmentation and balancing strategies, were applied during training. Model performance was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), sensitivity, F1-score, and precision. SHAP analysis was performed to identify important predictive features. Results: The tabular XGBoost model achieved the best predictive performance with an AUC of 0.95 and a sensitivity of 0.50, while five-fold cross-validation yielded an AUC of 0.79 and a sensitivity of 0.48. Among image-based models, ResNet50 achieved the highest performance, with an AUC of 0.76 and a sensitivity of 0.40. The multimodal model achieved an AUC of 0.57 and a sensitivity of 0.20. SHAP analysis identified cannulation time, ampulla type, and age as prominent features associated with PEP prediction. Conclusions: This exploratory study suggests that structured clinical data currently provide stronger predictive signals for PEP prediction than the available image and multimodal data within this limited cohort. The relatively low occurrence of PEP contributed to class imbalance despite mitigation strategies. Future multicenter studies with larger datasets, improved image availability, synthetic data generation, and advanced multimodal fusion techniques may improve predictive performance and clinical applicability. Full article

The integration of Generative Artificial Intelligence (GenAI) in higher education has opened new possibilities for optimizing administrative and academic services, particularly in contexts characterized by high-demand communication processes. Within the framework of service science, this study addresses the challenge of efficiently managing high volumes of email inquiries in a university master’s program, aiming to improve service quality and operational efficiency. The study examines the implementation of GenAI-based assistants, specifically NotebookLM and custom Gem AI assistants, trained in regulatory, curricular, and historical data from the University Master’s in Teacher Training at Rey Juan Carlos University. A mixed analytical approach is adopted, combining elements of data science to quantify efficiency gains and service science to analyze organizational and service-related transformations. The implementation of GenAI assistants contributes to improved response times, enhanced accuracy of information provided, and a reduction in administrative workload. The results suggest that GenAI can support the scalability and quality of academic administrative services when integrated within a structured service framework. However, its effective adoption requires careful consideration of ethical, organizational, and governance dimensions to ensure sustainable and responsible implementation. Full article

In single-phase PWM rectifiers, due to the inherent time-varying characteristics of the source voltage and current as well as the periodic operation of the converter bridge, the instantaneous input power on the AC side inevitably exhibits a twice-fundamental-frequency pulsation. This phenomenon consequently generates a double-line-frequency (100 Hz) voltage ripple on the DC-link capacitor, which causes an inherent contradiction in conventional voltage outer-loop control between steady-state ripple suppression and dynamic response speed. To address this issue, this paper proposes a control strategy based on an Adaptive Time-Delayed Feedforward Neural Network (Adaptive TD-FNN). The proposed method explicitly introduces the delayed voltage error of half a ripple period into the network state input, thereby achieving time-domain decoupling of the 100 Hz low-frequency disturbance. In addition, a physics-driven training framework is constructed by integrating the rectifier’s discrete difference equation, thereby strengthening the network’s capacity to learn the dynamic characteristics of the system. On this basis, a dynamic adaptive smoothness-weight penalty mechanism is designed to adjust the weighting factor of the current command smoothness constraint in the loss function according to the system operating state. Specifically, the penalty weight is increased under steady-state conditions to suppress command oscillations caused by ripple disturbances, while it is rapidly reduced during load or grid-voltage transients to release the network’s transient optimization capability. Simulation and experimental results show that the proposed Adaptive TD-FNN controller can simultaneously achieve smooth steady-state current command output and fast dynamic voltage regulation without introducing additional complex digital notch-filtering algorithms. Compared with conventional dual-loop control, the proposed strategy reduces the total harmonic distortion (THD) of the grid-side input current from 8.45% to 3.42%, satisfying grid-connected power quality requirements. Meanwhile, under large load transients and grid-voltage disturbance conditions, the DC-link voltage recovery time is about 40 ms, verifying the comprehensive advantages of the proposed method in ripple suppression, dynamic response, and operating-condition adaptability. Full article