Search Results (8,184)

Fire accidents in road tunnels can cause a significant number of fatalities and severe damage to tunnel structures. The tunnel European directive applies to the trans-European road network and requires the use of active smoke control systems in most tunnels longer than 1000 m. Research has investigated whether shorter tunnels without active smoke control systems are safe. If smoke contaminates the lower layer where people evacuate, it can impair visibility. This disturbs egress and may cause intoxication and, eventually, death. The FireFoam computer code was applied to the Memorial Tunnel fire ventilation tests for validation. This work investigates the effect of varying the heat release rate (HRR), ranging from 6 to 100 MW, under a wind velocity of 0.77 m/s and in the absence of wind. Results show that high HRR moves the start of lower layer smoke contamination closer to the fire source, reducing the distance from 390 m at 14 MW to as close as 210 m at 100 MW. An analytical model was developed to predict the distance from the fire source where smoke can contaminate the lower layer and was subsequently improved to account for HRR variation. Full article

Based on lessons learned from the Fukushima Daiichi Nuclear Power Plant accident caused by the 2011 off the Pacific coast Tohoku Earthquake, and the subsequent tsunamis, Japanese utilities have been upgrading their tsunami countermeasures. To understand the residual risk from beyond-design-basis events, it is important to assess seismic and tsunami risks independently while also recognizing how a plant’s risk profile changes when these events occur concurrently. This study developed a framework for a multi-hazard probabilistic risk assessment (PRA) to evaluate risks to nuclear power plants (NPPs) resulting from the superposition of earthquake and tsunami events. The framework is proposed on the assumption that the targeted plant has previously conducted single-hazard PRAs for both earthquakes and tsunamis. This study presents an approach to define the scope of risk assessment for the superposition of earthquake and tsunami events, based results from single-hazard PRAs for each hazard. It provides an analytical framework for superposition scenario analysis and a simplified method for multi-hazard assessment using data from single-hazard assessments. Moreover, a series of steps for the multi-hazard fragility assessment of superposed earthquake and tsunami events are proposed, clarifying the relationship between superposed impacts and the damaged parts and damage modes, accompanied by illustrative examples. Full article

Vehicle crush boxes are one of the safety elements used in vehicles to minimize damage that may occur during an accident. The task of crush boxes is to absorb the energy which is generated during an accident. In this study, peak force, energy absorption and specific energy absorption values of cylindrical composite crush boxes, to which 0.25% and 0.50% graphene was added, were experimentally investigated with hydrothermal aging. The composite crush boxes were produced with vacuum infusion method. Glass, aramid and carbon fibers and their hybridizations were used as fibers. During hybridization, the winding order of the fibers was changed from inside to outside. The parameters for hydrothermal aging were selected as 500 h and 1000 h at 60 °C. The highest energy absorption value was obtained in the carbon fiber-reinforced sample CFRPG1H2 with 0.25% graphene-added epoxy resin matrix, aged for 1000 h. The lowest peak strength was observed in the aramid fiber-reinforced sample AFRPG2H2 with 0.50% graphene-added epoxy resin matrix, hydrothermally aged for 1000 h. It was observed that increasing the graphene addition rate reduced the negative effects on aging. It was determined that increasing the graphene ratio by 0.25% had an effect on aging. Full article

Intelligent Transportation Systems (ITS) have progressively incorporated machine learning to optimize traffic efficiency, enhance safety, and improve real-time decision-making. However, the traditional centralized machine learning (ML) paradigm faces critical limitations regarding data privacy, scalability, and single-point vulnerabilities. This study explores FL as a decentralized alternative that preserves privacy by training local models without transferring raw data. Based on a systematic literature review encompassing 39 ITS-related studies, this work classifies applications according to their architectural detail—distinguishing systems from models—and identifies three families of federated learning (FL) frameworks: privacy-focused, integrable, and advanced infrastructure. Three representative frameworks—Federated Learning-based Gated Recurrent Unit (FedGRU), Digital Twin + Hierarchical Federated Learning (DT + HFL), and Transfer Learning with Convolutional Neural Networks (TFL-CNN)—were comparatively analyzed against a client–server baseline to assess their suitability for ITS adaptation. Our qualitative, architecture-level comparison suggests that DT + HFL and TFL-CNN, characterized by hierarchical aggregation and edge-level coordination, are conceptually better aligned with scalability and stability requirements in vehicular and traffic deployments than pure client–server baselines. FedGRU, while conceptually relevant as a meta-framework for coordinating multiple organizational models, is primarily intended as a complementary reference rather than as a standalone architecture for large-scale ITS deployment. Through application-level evaluations—including traffic prediction, accident detection, transport-mode identification, and driver profiling—this study demonstrates that FL can be effectively integrated into ITS with moderate architectural adjustments. This work does not introduce new experimental results; instead, it provides a qualitative, architecture-level comparison and adaptation guideline to support the migration of ITS applications toward federated learning. Overall, the results establish a solid methodological foundation for migrating centralized ITS architectures toward federated, privacy-preserving intelligence, in alignment with the evolution of edge and 6G infrastructures. Full article

Predicting the severity of traffic accidents remains challenging due to the limited ability of existing methods to extract deep semantic information from unstructured accident narratives, as traditional approaches typically depend on structured data alone. This study proposes a severity prediction approach enhanced by semantic risk reasoning derived from large language models (LLMs). A prompt-engineering template is designed to guide LLMs in extracting proxy semantic features from accident descriptions, forming an enriched feature set that incorporates causal logic. These semantic features are fused with traditional structured features through three integration strategies—direct feature concatenation, optimized feature selection, and model-level fusion. Experiments based on 4013 accident records from expressways in Yunnan Province, China, demonstrate that models using LLM-derived semantic features significantly outperform those relying solely on structured features. Notably, the LightGBM model utilizing semantic features within a balanced learning framework achieves a severe accident recall of 77.8%. While model-level fusion proves optimal for XGBoost (improving Macro-F1 to 0.6356), we identify a “feature dilution” effect in other classifiers, where high-quality semantic reasoning is compromised by low-quality structured noise. These findings indicate that the proposed approach effectively enhances the identification of high-risk accidents and offers a novel semantic-aware solution for traffic safety management. Furthermore, the obtained results provide actionable insights for traffic management agencies to optimize emergency response resource allocation and formulate targeted accident prevention strategies. Full article

This paper presents a practical and cost-effective method for in-motion estimation of a vehicle’s CoG position in all three directions by measuring accelerations during two types of maneuvers: braking (longitudinal and vertical CoG estimation) and cornering (lateral and vertical CoG estimation). The proposed method’s main advantage is that it does not require knowledge of vehicle characteristics, such as mass distribution, suspension geometry, or inertia parameters. It relies solely on the known distances between the sensors and their positions relative to a defined reference point on the vehicle. To validate the developed method, experimental tests were conducted on a prototype vehicle, varying the load conditions for the proposed driving scenarios. The CoG position obtained from dynamic maneuvers was compared with reference values derived from static measurements. The results showed that the proposed method could estimate the CoG position with an average error of 3% in the longitudinal direction, a maximum error of 12% in the lateral direction, and a maximum error of 14% in the vertical direction. Full article

In recent years, despite the continuous improvement of China’s construction safety management systems and the adoption of advanced technologies, safety accidents remain frequent. This shift highlights the growing importance of human factors in construction safety. As the main labor force, the new generation of construction workers differs significantly from previous generations in values and motivation, reducing the effectiveness of traditional safety management models. This study investigates the direct effect of transformational leadership on the safety behavior of new-generation construction workers. Using survey data collected from construction enterprises in Guangdong Province, China, and applying structural equation modeling (SEM), the results reveal that transformational leadership has a significant positive impact on safety behavior. All four dimensions—idealized influence, inspirational motivation, idealized influence (charisma) and individualized consideration—positively influence both safety compliance and participation, with inspirational motivation exerting the strongest effect (β = 0.509 for compliance; β = 0.446 for participation). These findings indicate that leaders who articulate a compelling shared vision can effectively internalize safety norms and motivate proactive safety participation. This study enriches theoretical understanding of safety leadership mechanisms and provides practical guidance for construction enterprises to enhance safety performance through cultivating transformational leadership among managers. Full article

A strong safety culture is essential for managing human factors in complex systems and constitutes a strategic resource for supporting the sustainable operation of organizations. However, conventional approaches remain limited by unclear conceptual boundaries and a lack of mechanisms linking safety culture with other organizational safety elements. To address these gaps, this study develops a sustainable safety culture construction method grounded in accident causation theory. Using the 24Model, we establish a concise “culture–system–ability–acts” framework that operationalizes the pathways through which safety culture shapes organizational safety performance. The method integrates four components: conceptual clarification of safety culture, quantitative assessment, factor identification based on the 24Model, and Bayesian network analysis to quantify interdependencies among culture, systems, ability, and acts. Empirical evidence from coal mining enterprises shows that safety culture influences safety performance indirectly by shaping system implementation quality, workers’ safety ability, and safety-related actions. Enhancing “demand of safety training” substantially mitigated system deficiencies related to ineffective implementation of procedures, failure in enforcing procedures, lack of qualifications, and insufficient supervision. Improved training also strengthened workers’ knowledge of accident cases, consequences of violations, and technical standards, thereby reducing competence-related gaps and promoting more consistent safety supervision behaviors. Sensitivity analysis highlights the importance of reinforcing “safety responsibilities of line departments” and improving the dissemination of safety knowledge, particularly accident case knowledge. Overall, the findings empirically validate the dynamic “culture–system–ability–acts” transmission mechanism of the 24Model and provide a structured, quantitative pathway for advancing sustainable safety culture development. Full article

The emerging urbanization and the extensive increase of the transportation sector are responsible for the significant increase in carbon dioxide emissions. Therefore, replacing traditional cars with Electric Vehicles (EVs) is a promising solution, offering a clearer alternative. EVs are becoming more and more well-known and are being quickly used worldwide. However, the exponential rise in EV sales has also raised a number of issues, which are becoming important and demanding. These challenges include the need of driving security, the battery degradation, the inadequate infrastructure for charging EVs, and the uneven energy distribution. In order for EVs to reach their full potential, intelligent systems and innovative technologies need to be introduced in the field of EVs. This is where business intelligence (BI) can be employed, along with artificial intelligence (AI), data analytics, and machine learning. In this paper, we provide a comprehensive survey on the use of BI strategies in the EV transportation sector. We first introduce the EVs and charging station technologies. Then, research works on the application of BI and data analysis techniques in EV technology are reviewed to further understand the challenges and open issues for the research and industry community. Moreover, related works on accident analysis, battery health prediction, charging station analysis, intelligent infrastructure, locating charging stations analysis, and autonomous driving are investigated. This survey systematically reviews 75 peer-reviewed studies published between 2020 and 2025. Finally, we discuss the fundamental limitations and the future open challenges in the aforementioned topics. Full article

The energy efficiency and hover stability of unmanned aerial vehicles are critical factors, since improper battery utilization and unstable control are major sources of operational failures and accidents. The proportional–integral–derivative (PID) controller, which is applied in approximately 97% of multirotor unmanned aerial vehicle (UAV) systems, is widely used due to its simplicity; however, it is sensitive to external disturbances and often fails to ensure optimal energy utilization, resulting in reduced flight time. Therefore, the experimental investigation of advanced control methods in a real physical environment is well justified. The objective of the present research is the comparative evaluation of seven control strategies—PID, linear quadratic controller with integral action (LQI), model predictive control (MPC), sliding mode control (SMC), backstepping control, fractional-order PID (FOPID), and H∞ control—using a single-degree-of-freedom drone test platform in a MATLAB R2023b-Arduino hardware-in-the-loop (HIL) environment. Although the theoretical advantages and model-based results of the aforementioned control methods are well documented, the number of real-time comparative HIL experiments conducted under identical physical conditions remains limited. Consequently, only a small amount of unified and directly comparable experimental data is available regarding the performance of different controllers. The measurements were performed at a reference height of 120 mm under disturbance-free conditions and under wind loading with a velocity of 10 km/h applied at an angle of 45°. The controller performance was evaluated based on hover accuracy, settling time, overshoot, and real-time measured power consumption. The results indicate that modern control strategies provide significantly improved energy efficiency and faster stabilization compared to the PID controller in both disturbance-free and wind-loaded test scenarios. The investigations confirm that several advanced controllers can be applied more effectively than the PID controller to enhance hover stability and reduce energy consumption. Full article

Nonlinear single-input single-output (SISO) systems operating under parametric uncertainty often exhibit bifurcations, multistability, and deterministic chaos, which significantly limit the effectiveness of classical linear, adaptive, and switching control methods. This paper proposes a novel synthesis framework for self-organizing control systems based on catastrophe theory, specifically within the class of elliptic catastrophes. Unlike conventional approaches that stabilize a predefined system structure, the proposed method embeds the control law directly into a structurally stable catastrophe model, enabling autonomous bifurcation-driven transitions between stable equilibria. The synthesis procedure is formulated using a Lyapunov vector-function gradient–velocity method, which guarantees aperiodic robust stability under parametric uncertainty. The definiteness of the Lyapunov functions is established using Morse’s lemma, providing a rigorous stability foundation. To support practical implementation, a data-driven parameter tuning mechanism based on self-organizing maps (SOM) is integrated, allowing adaptive adjustment of controller coefficients while preserving Lyapunov stability conditions. Simulation results demonstrate suppression of chaotic regimes, smooth bifurcation-induced transitions between stable operating modes, and improved transient performance compared to benchmark adaptive control schemes. The proposed framework provides a structurally robust alternative for controlling nonlinear systems in uncertain and dynamically changing environments. Full article

To identify and evaluate the failure issues in the livestock manure-to-biomethane supply chain, this study employs a Bayesian network approach with three inference analysis methods: diagnostic analysis, sensitivity analysis, and maximum causal chain inference. First, the main hazard categories affecting the failure of the supply chain are identified, establishing risk indicators for feedstock collection, pretreatment, anaerobic digestion, purification and upgrading, transportation, and biomethane end-use. Then, the half-interval method and possibility superiority comparison are used to calculate and rank the severity of related accidents, obtaining the severity ranking of secondary indicators as well as the severity ranking of work items and risk items. Finally, Bayesian forward inference is applied to investigate the failure probability of the supply chain, combined with backward inference to identify the risk factors most likely to cause supply chain failures and trace the formation of failure hazards. The Bayesian sensitivity analysis method is ultimately applied to determine the key hazards affecting supply chain failures and the correlations between accident hazards, followed by validation. The results show that the failure probability of the supply chain through causal inference is approximately 54.76%, indicating relatively high failure risk. The three factors with the highest posterior probabilities are mechanical stirring failure C₃ (88.11%), corrosion-induced ammonia leakage poisoning D6, and equipment explosion caused by excessive pressure due to overheating during dehumidification heating D₉, which are the hazards most likely to cause failures in the supply chain. Improper operations and the toxicity of related chemicals are key hazards leading to supply chain failures, with the correlation between accident hazards presented as a hazard chain by integrating severity and accident probability, and the key risk points in the supply chain are identified. Full article

Traditional automated monitoring systems adopted for Intersection Traffic Control still face challenges, including high costs, maintenance difficulties, insufficient coverage, poor multimodal data integration, and limited traffic information analysis. To address these issues, the study proposes a sovereign AI-driven Smart Transportation governance approach, developing a mobile AI solution equipped with multimodal perception, task decomposition, memory, reasoning, and multi-agent collaboration capabilities. The proposed system integrates computer vision, multi-object tracking, natural language processing, Retrieval-Augmented Generation (RAG), and Large Language Models (LLMs) to construct a Pipeline-based Traffic Analysis System (PTAS). The PTAS can produce real-time statistics on pedestrian and vehicle flows at intersections, incorporating potential risk factors such as traffic accidents, construction activities, and weather conditions for multimodal data fusion analysis, thereby providing forward-looking traffic insights. Experimental results demonstrate that the enhanced DuCRG-YOLOv11n pre-trained model, equipped with our proposed new activation function $\beta silu$, can accurately identify various vehicle types in object detection, achieving a frame rate of 68.25 FPS and a precision of 91.4%. Combined with ByteTrack, it can track over 90% of vehicles in medium- to low-density traffic scenarios, obtaining a 0.719 in MOTA and a 0.08735 in MOTP. In traffic flow analysis, the RAG of Vertex AI, combined with Claude Sonnet 4 LLMs, provides a more comprehensive view, precisely interpreting the causes of peak-hour congestion and effectively compensating for missing data through contextual explanations. The proposed method can enhance the efficiency of urban traffic regulation and optimizes decision support in intelligent transportation systems. Full article

Automotive dashboard cameras are widely used to record driving events and often serve as critical evidence in accident investigations and insurance claims. However, the availability of free and low-cost editing tools has increased the risk of video tampering, underscoring the need for reliable methods to verify video authenticity. Temporal tampering typically involves manipulating frame order through insertion, deletion, or duplication. This paper proposes a computationally efficient framework that transforms high-dimensional video into compact one-dimensional temporal signals and learns tampering patterns using a shallow one-dimensional convolutional neural network (1D-CNN). Five complementary features are extracted between consecutive frames: frame-difference magnitude, structural similarity drift (SSIM drift), optical-flow mean, forward–backward optical-flow consistency error, and compression-aware temporal prediction error. Per-video robust normalization is applied to emphasize intra-video anomalies. Experiments on a custom dataset derived from D²-City demonstrate strong detection performance in single-attack settings: 95.0% accuracy for frame deletion, 100.0% for frame insertion, and 95.0% for frame duplication. In a four-class setting (non-tampered, insertion, deletion, duplication), the model achieves 96.3% accuracy, with AUCs of 0.994, 1.000, 0.997, and 0.988, respectively. Efficiency analysis confirms near real-time CPU inference (≈12.7–12.9 FPS) with minimal memory overhead. Cross-dataset tests on BDDA and VIRAT reveal domain-shift sensitivity, particularly for deletion and duplication, highlighting the need for domain adaptation and augmentation. Overall, the proposed multi-feature 1D-CNN provides a practical, interpretable, and resource-aware solution for temporal tampering detection in dashcam videos, supporting trustworthy video forensics in IoT-enabled transportation systems. Full article

Research into Remotely Piloted Aircraft Systems (RPASs) has expanded rapidly, yet the competencies, knowledge, skills, and other attributes (KSaOs) required of RPAS pilots remain comparatively underexamined. This review consolidates existing studies addressing human performance, subject matter expertise, training practices, and accident causation to provide a comprehensive account of the KSaOs underpinning safe civilian and commercial drone operations. Prior research demonstrates that early work drew heavily on military contexts, which may not generalize to contemporary civilian operations characterized by smaller platforms, single-pilot tasks, and diverse industry applications. Studies employing subject matter experts highlight cognitive demands in areas such as situational awareness, workload management, planning, fatigue recognition, perceptual acuity, and decision-making. Accident analyses, predominantly using the human factors accident classification system and related taxonomies, show that skill errors and preconditions for unsafe acts are the most frequent contributors to RPAS occurrences, with limited evidence of higher-level latent organizational factors in civilian contexts. Emerging research emphasizes that RPAS pilots increasingly perform data-collection tasks integral to professional workflows, requiring competencies beyond aircraft handling alone. The review identifies significant gaps in training specificity, selection processes, and taxonomy suitability, indicating opportunities for future research to refine RPAS competency frameworks and support improved operational safety. Full article