Search Results (8,666)

This study investigates the structural misalignment between official wage benchmarks and actual market wages in the Republic of Korea to establish an independent, forestry-specific wage system essential for labor sustainability. Historically, the Republic of Korea forestry project costs have relied on construction industry benchmarks, leading to a “diverging hypothesis” where official rates fail to reflect the specialized risks and technical skills required in forest operations. To address this, a comprehensive wage survey was conducted in 2025 across 13 specialized forestry occupations. Utilizing a sampling frame of 7555 sites, 1044 units were selected via stratified sampling with square-root proportional allocation, ensuring a relative standard error (RSE) of 2.5%. The findings reveal that market wages consistently exceed construction benchmarks by 4.5% to 41.0%. The most significant disparities were observed in leadership and mechanized roles, reflecting substantial “risk–responsibility” and “skill premiums”. Furthermore, the study identifies a structural shift toward risk-transfer strategies, such as stumpage sales, in response to the Serious Accidents Punishment Act (SAPA). These results underscore the urgent need for a specialized wage framework to ensure safety and long-term resilience. Ultimately, such institutional refinement is a prerequisite for securing the high-quality human capital necessary for a sustainable circular bioeconomy. Full article

Same-level falls are the most frequent occupational accidents, yet traditional manual analysis of accident reports is labor-intensive and limits large-scale prevention strategies. In this pilot study, we aimed to evaluate the accuracy of using large language models (LLMs) to automate the classification of occupational accident text data without task-specific pretraining. We analyzed data from 2619 same-level-fall-related injury cases, using expert manual classification as the reference standard. Four models—GPT-4o mini, GPT-4.1 mini, GPT-4.1, and o4-mini—were compared using accuracy and Cohen’s kappa. The o4-mini model demonstrated the highest performance, showing statistical superiority in the complex “causal agent” category with 72.8% accuracy. For other classification tasks, the top models achieved accuracies of 82–92%, with Cohen’s kappa coefficients > 0.7, indicating substantial agreement with expert judgments. These findings suggest that LLMs can classify occupational accident text with substantial agreement with the expert-derived reference standard in this dataset. This automated approach enables efficient, high-frequency analysis of large datasets, offering a promising tool for large-scale occupational accident surveillance and screening. Full article

Road construction, inspection and maintenance are activities that often require workers near heavy equipment, traffic, and dangerous materials. This proximity to potential hazards along with the characteristics of highway and street work zones—transient and in restricted areas—increases the possibility of accidents and near-misses. Recent developments in remote-controlled technology can provide workers and inspectors with the ability to conduct activities from a safer distance. This paper aims to scan and evaluate several promising remote-controlled technologies that could be used to improve safety in highway and streets work zones. The technology scanning highlighted over twenty technologies in several levels of development that met this goal. Each technology was briefly evaluated not only based on safety features but also on productivity, data processing, and requirements for implementation. Finally, recommendations for implementation of selected technologies were provided. This consolidated review provides a unique and timely resource for researchers and practitioners. Full article

The Fukushima nuclear accident highlighted that evacuation-related psychosocial harm can outweigh direct radiation risks, underscoring the need to define the health impacts of chronic low-dose-rate (LDR) radiation and evidence-based thresholds for intervention. This study investigated the effects of continuous, postnatal LDR gamma irradiation (1.2 mGy/h, cumulative dose: 5 Gy) in male mice. While no changes in body weight, hippocampal neurogenesis, or major glial and neuronal populations were observed, persistent DNA damage (γ-H2AX foci) in dentate gyrus granule cells occurred in both irradiated male and female mice. Irradiated male mice developed anxiety-like behaviour, a phenotype not observed in a previously published study of female mice subjected to an identical irradiation protocol. Molecular profiling revealed two novel, dysregulated miRNA/mRNA axes in the hippocampus linking DNA damage to behaviour: a maladaptive miR-466i-5p/Tfcp2l1 pathway associated with genomic instability, and a potentially adaptive miR-101a-5p/BMP6 pathway promoting neuronal survival. Venn analysis further identified miR-124b-3p and novel-miR489-3p as conserved exposure biomarkers, altered in both the hippocampus and blood of irradiated animals. Our results show that a high cumulative dose of chronic LDR induces markedly less severe hippocampal pathology than has been reported for equivalent acute doses. These findings support the concept of dose-rate-dependent threshold dose and contribute to the evidence base for developing countermeasures following nuclear incidents or other radiation exposures. Full article

As the number of elderly individuals living alone increases, the risk of fall-related accidents correspondingly rises, underscoring the need for rapid fall detection systems. Because falls are difficult to predict in terms of location, detection systems must be deployed in a distributed manner, which in turn requires compact and low-power implementations. Unlike camera sensors, radar sensors do not raise privacy concerns and are not limited by line-of-sight constraints. Moreover, compared with wearable sensors, radar enables continuous monitoring without user intervention. However, prior radar-based approaches incur high computational complexity, leading to increased power consumption and larger hardware area, thereby necessitating efficient hardware design. This paper proposes a lightweight fall detection system based on continuous-wave (CW) radar and a binarized convolutional neural network (BCNN). Radar signals are preprocessed using short-time Fourier transform (STFT) to generate binary spectrograms, which are then fed into a BCNN-based classification network. The proposed system performs binary classification of five fall activities and seven non-fall activities with an accuracy of 96.1%. The preprocessing module and classification network were implemented as hardware accelerators and integrated with a microprocessor in a system-on-chip (SoC) architecture on a field-programmable gate array (FPGA). Compared with the software implementation, the proposed hardware achieved speedups of 387.5× and 86.7× for the preprocessing and classification modules, respectively. Furthermore, the overall system processing time was 2.58 ms, corresponding to an 89.5× speedup over the software baseline. Full article

This study evaluates the risk reduction performance of prescriptive technical codes applied to hydrogen refueling stations using a Layer of Protection Analysis (LOPA) approach. A representative accident scenario involving high-pressure hose rupture at the dispenser was selected as the initiating event, and the initiating event frequency was determined based on CCPS guidelines. The target mitigated event likelihood (TMEL) was set to $1.0\times {10}^{-6}$/year, resulting in a required risk reduction factor (RRF) of $1.0\times {10}^4$. Safety devices specified in the Korean Gas Safety (KGS) Codes were identified as independent protection layers (IPLs), and their probability of failure on demand (PFD) values were assigned based on commonly accepted LOPA data. The combined PFD of the identified IPLs was estimated to be $1.0\times {10}^{-5}$, leading to a mitigated event likelihood of $1.0\times {10}^{-7}$/year, which satisfies the predefined TMEL. These results indicate that the prescriptive technical codes can provide a certain level of quantitative risk reduction when all required safeguards operate as assumed. However, the analysis also reveals structural limitations associated with independence assumptions, potential common cause failures, and maintenance conditions. The findings suggest that integrating functional safety concepts and systematic risk assessment with prescriptive codes could enhance the reliability of safety management for hydrogen refueling stations. Full article

The development of accident-tolerant fuels has significantly enhanced the safety of fission reactors. The TiNbZrO alloy system has garnered considerable attention due to its excellent mechanical properties and outstanding irradiation resistance. Its unique compositional design enables effective suppression of irradiation-induced defect formation. In this study, TiNbZrO thin films are fabricated via radio-frequency magnetron sputtering and irradiated with 50 keV He ions to fluences of 5 × 10¹⁶, 1 × 10¹⁷, and 2 × 10¹⁷ ions/cm². The microstructural evolution before and after irradiation is characterized by Transmission Electron Microscopy (TEM) and Grazing Incidence X-ray Diffraction (GIXRD), and the changes in mechanical properties are evaluated by nanoindentation. With increasing irradiation fluence, the average size of He bubbles increases from 1.10 nm to 2.06 nm, the number density decreases from 5.27 × 10²⁴ m⁻³ to 1.39 × 10²⁴ m⁻³, and the swelling rate rises from 0.37% to 0.64%. Although significant irradiation hardening is observed in all samples, the maximum hardening rate reaches only 31.91%, a value substantially lower than that reported for many conventional nuclear materials. This demonstrates the superior irradiation resistance of TiNbZrO thin films. The superior irradiation resistance of TiNbZrO thin films stems from two synergistic effects: high-entropy lattice distortion suppresses atomic diffusion, while oxygen complexes pin defects. Full article

Road traffic safety is central to socially resilient and sustainable cities, yet many middle-income countries lack harmonized subnational data on exposure, infrastructure, and enforcement. This study examines whether routinely available demographic composition can serve as a practical structural baseline for provincial traffic accident rates and as a diagnostic layer for richer safety models. Using official province–year data from Türkiye (2008–2019 and 2022–2024; n = 1215), demographic shares by sex, education, and age were treated as compositional inputs and transformed using isometric log-ratio (ILR) methods, with GDP per person included as a scalar covariate. A Tabular Residual Network (ResNet) was trained on the historical panel and evaluated on a post-period calibration/evaluation window (2022–2024), which was used for checkpoint selection and seed screening rather than as an independent held-out test set. Among the evaluated specifications, the ResNet seed-ensemble achieved the strongest performance on the 2022–2024 calibration/evaluation period (R² = 0.5717), outperforming the best single-seed model (R² = 0.5539), a province-specific last-value-carried-forward temporal heuristic based on 2019 values (R² = 0.4779), tree-based tabular benchmarks (Random Forest: R² = 0.1328; XGBoost: R² = 0.0706), and pooled statistical reference models (linear: R² = 0.1375; negative binomial: R² = 0.0686; Poisson: R² = −0.0634). Year-wise diagnostics indicated gradual temporal drift, suggesting that periodic recalibration or the inclusion of additional policy-relevant covariates is needed to preserve calibration. Overall, ILR-based compositional geodemography provides a scalable and interpretable baseline for traffic safety monitoring and prioritization in data-constrained settings. Full article

Under the global energy transition, liquefied petroleum gas (LPG) remains an important transitional fuel. However, persistent safety risks in storage and transportation continue to limit its sustainable development. This study aims to evaluate the sustainability of China’s LPG storage and transportation system and identify practical improvement pathways. A “1+4” C-STSM multidimensional framework was developed by combining accident fault-tree analysis, comparative review of domestic and international standards, and a systematic assessment of storage, transportation, monitoring, and safety technologies. The results show that the sustainability of LPG systems depends on the coordinated performance of infrastructure, transportation, monitoring, and safety barriers across the full supply chain. China has made progress in engineering facilities and safety management, but still faces weaknesses in intrinsic safety, barrier integrity, intelligent monitoring, and life-cycle governance. The main gap with international advanced practice lies in insufficient system integration rather than the lack of basic technologies. Improving LPG sustainability requires a coordinated pathway that combines safer infrastructure, intelligent monitoring, stronger barrier management, and better regulatory coordination. Such an approach can enhance industrial safety while supporting low-loss, low-emission energy transition. Full article

The primary aim of wildlife rehabilitation centres is the release of treated animals back into the wild after treatment. Data collected during rehabilitation can provide valuable insights into current trends and can help with conservation strategies aiming to reduce human-related negative impact. This review analyses records from past and currently operating wildlife rescue and rehabilitation centres in Lithuania over a 23-year period. Data were compiled for 7847 individual animals representing 216 species of birds, mammals, and reptiles. The study evaluates patterns of seasonal admission, taxonomic composition, known causes of injury or admission, and rehabilitation outcomes. The results showed that 83% of cases involved birds, mammals comprised 16%, and reptiles were recorded only infrequently (0.52%). Admissions peaked in summer, comprising 42% of all cases. Injuries of unknown origin were the most common (55%). Among cases with identified causes, a substantial proportion were associated with human activities, including road accidents (5% of all cases), collisions with anthropogenic structures (4%), and attacks by domestic cats or dogs (3%). The presence of a specialized rehabilitation centre, together with active public involvement in the rescue of injured wildlife, contributes to release rates reaching approximately 30% of admitted animals. Datasets such as those analyzed in this study may contribute to improved preparedness for managing unavoidable human–wildlife interactions in the future. Full article

Road traffic injuries represent one of the most critical public health challenges in the Gulf region. Predicting traffic accident severity is therefore a critical component of evidence-based road safety management. In this study, we develop machine learning frameworks for predicting traffic accident severity using Qatar’s national dataset (2020–2025), addressing extreme class imbalance and interpretability. A dataset of 588,023 accident records was systematically preprocessed from 1,000,500 raw reports. We compare three approaches: multi-class (four severity levels), binary (Safe vs. Severe), and cascaded two-stage (combining both). Six classifiers were evaluated across two encoding methods and three balancing strategies. Systematic hyperparameter tuning with 5-fold stratified cross-validation was performed for all models. The binary LightGBM classifier achieved BA = 71.04%, AUC-ROC = 0.772, Sensitivity = 61.03%, and Specificity = 81.05%, demonstrating superior performance over multi-class approaches. Temporal validation on 2025 data (trained on 2020–2024 data) supported good temporal generalization. Analysis of 10,000 test instances identified the time period as the dominant predictor of accident severity. The binary LightGBM framework provides an interpretable and effective approach for severe accident identification and risk prioritization, with SHAP findings supporting targeted temporal enforcement and pedestrian safety as evidence-based policy priorities. Full article

Background: Mandibular fractures constitute a significant proportion of maxillofacial trauma resulting from traffic accidents and present valuable information about the severity of the trauma mechanism. The aim of this study was to evaluate the demographic characteristics, fracture patterns, and accompanying injuries of mandibular fractures resulting from traffic accidents. Methods: A retrospective examination was made of 94 patients who presented for forensic medicine evaluation following a traffic accident between 1 January 2019 and 31 December 2024 and were determined with mandibular fracture. The demographic data, accident characteristics, localization of the mandibular fracture, number of fractures, displacement status, and accompanying injuries were analyzed. Results: The analyzed cases comprised 68.1% males and 31.9% females, with a mean age of 29.27 ± 14.34 years. The mandibular fractures were displaced in 52.1% of cases, and closed in 98.9%. The fracture regions were determined to most often be the ramus (32.9%) and the condyle (32.9%). A single fracture was present in 54.9% of cases and multiple fractures in 45.1%. A significant correlation was seen between ramus fractures and male gender, driver status, and concomitant systemic injuries, whereas no significant relationship was found between some fracture types and the demographic and accident-related variables. Conclusions: Mandibular fractures resulting from traffic accidents may represent relatively high-energy trauma mechanisms, and certain fracture patterns may occur together with multiple and systemic injuries. The localization and characteristics of mandibular fractures present important clues about the biomechanics of the trauma and a holistic approach is required in the forensic medicine evaluation. Full article

This study proposes an improved tree-augmented Bayesian network (TAN-BN) method for analyzing the severity of ship collision accidents by introducing the information contribution rate (ICR) for edge orientation and flexible filtering constraints for structure optimization. Based on 634 ship collision accident reports, a Bayesian network covering accident attributes and causal factors was constructed. The results show that the improved model achieved an overall AUC of 0.864, higher than that of the traditional TAN model (0.827). Mutual information analysis identified ship length as the factor most strongly associated with accident severity, with a mutual information value of 0.0868. Sensitivity analysis based on true risk impact (TRI) further showed that ship length, time, and ship type were the most influential factors, with average TRI values of 19.4%, 8.8%, and 7.2%, respectively. The proposed model effectively captures the dependency relationships between accident severity and multiple influencing factors and can provide quantitative support for risk warning and accident prevention in maritime traffic safety. Full article

Background/Objectives: Injury mechanism and intent are key determinants of patient outcomes in the emergency department, yet their combined effects remain insufficiently understood. Emergency department disposition after injury may differ according to mechanism and intent and may be further influenced by specific mechanism–intent combinations. This study aimed to evaluate the associations of injury mechanism, intent, and their interaction with emergency department disposition and injury severity measured using the International Classification of Diseases-based Injury Severity Score (ICISS). Methods: We conducted a retrospective analysis of injury-related emergency department visits recorded between 1 January 2019 and 31 December 2023. Eligible visits included those with valid arrival and departure timestamps and complete disposition data; records with missing key variables or implausible length of stay were excluded. A total of 1,029,875 visits were analyzed. The primary outcome was emergency department disposition, categorized as discharge, admission, or transfer. Multinomial logistic regression was used to estimate relative risk ratios, with discharge as the reference category, and to derive predicted probabilities for selected mechanism–intent combinations. Injury severity was assessed using ICISS and modeled with injury mechanism, intent, their interaction, and prespecified covariates. Results: Of all visits, 69.9% resulted in discharge, 24.3% in admission, and 5.8% in transfer. Compared with traffic accidents, the highest likelihood of admission was observed in suffocation, drowning, and poisoning injuries. Transfer was more frequent in drowning, suffocation, penetrating injuries, and poisoning. Self-harm was associated with increased risks of both admission and transfer compared with unintentional injuries. Interaction analyses showed that certain combinations, particularly poisoning with self-harm and suffocation with self-harm, were associated with substantially higher risks than either factor alone. Predicted probabilities further highlighted high-risk combinations, including markedly elevated admission probabilities in self-harm-related poisoning and suffocation, and increased transfer probability in unintentional drowning. Injury mechanism, intent, and selected interactions were also significantly associated with ICISS-based injury severity. Conclusions: Injury mechanism and intent are independently associated with emergency department disposition and injury severity, with additional risk amplification observed for specific combinations. These findings suggest that mechanism–intent combinations may serve as clinically useful risk indicators in emergency department triage and decision-making, supporting improved risk stratification and system-level coordination. Full article

Accurate forecasting of fatal construction accidents is critical for proactive safety management; however, accident time series exhibit strong non-stationarity, nonlinear dynamics, and multi-scale temporal patterns that challenge conventional models. This study proposes a hybrid STL–BiGRU framework that integrates Seasonal–Trend decomposition using Loess (STL) with a Bidirectional Gated Recurrent Unit (BiGRU) network to deliver robust and interpretable forecasts tailored to construction safety needs. STL first decomposes the original monthly accident series (January 2012–December 2024, OSHA) into trend, seasonal, and residual components, reducing structural complexity and mitigating non-stationarity. Independent BiGRU models are then trained on each component to capture bidirectional temporal dependencies, and final forecasts are reconstructed through component aggregation. Comparative experiments against Gated Recurrent Units (GRUs), Long Short-Term Memory (LSTM), Recurrent Neural Networks (RNNs), Support Vector Regression (SVR), Autoregressive Integrated Moving Average (ARIMA), and their STL-enhanced variants demonstrate that the proposed STL–BiGRU model achieves superior performance across both short-term and medium-term horizons. The model achieves the lowest error levels, with a short-term Root Mean Squared Error (RMSE) of 6.8522 and a medium-term RMSE of 7.0568, and shows consistent improvements in Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Results indicate that multi-scale decomposition combined with bidirectional deep learning provides a practical, forward-looking tool. It helps regulators and contractors anticipate high-risk periods, optimize resource allocation, and reduce fatal accidents through targeted preventive measures. Full article