IF 3.4 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH

安全科学与韧性(英文)

Pub Date : 2025-10-02 DOI: 10.1016/j.jnlssr.2025.100267

Guokang Xu , Jianchuan Yin , Nini Wang , Zeguo Zhang

Maritime transportation plays a crucial role in global economic trade. However, maritime accidents occur frequently, posing significant threats to the safety of seafarers. In search and rescue scenarios for man-overboard, unmanned aerial vehicles (UAVs) are gradually replacing manned aircraft and helicopters. Besides, detecting man-overboard is challenging because of their small pixel size, weak signals, and indistinct features on the ocean surface. Furthermore, existing detectors struggle to strike a balance between lightweight design for UAVs and detection accuracy. To address this issue, the novel Man-overboard Detection Transformer (MOB-DETR) is proposed. On the one hand, the Token Enhancement layer is introduced, which conducts fine-grained filtering of spatial and channel dimensions, reducing redundant encoding caused by background queries. On the other hand, the Effusion Fusion Module, based on the RepViT Block, is proposed, effectively eliminating computational redundancy by decoupling the interaction mechanisms between spatial and channel dimensions. Additionally, to fill the existing gap in benchmark datasets for detecting man-overboard, the ManOverboard benchmark dataset has been established. In the experimental validation phase, MOB-DETR is conducted on ManOverboard and SeaDronesSeev2. Ablation experiments show that MOB-DETR achieves 11.7 % better lightweight performance and 14.4 % higher

A P_{small}

than baselines. Comparison experiments on ManOverboard and SeaDronesSeev2 validate its effectiveness, offering an efficient solution for man-overboard detection. Overall, this research not only advances man-overboard detection but also significantly enhances the resilience of maritime transportation, ultimately protecting seafarers' lives and ensuring the reliability of the world's essential trade routes.

海上运输在全球经济贸易中起着至关重要的作用。然而，海上事故频发，对海员的安全构成重大威胁。在落水人员搜救领域，无人机正逐步取代有人驾驶飞机和直升机。此外，探测落水人员具有挑战性，因为它们的像素尺寸小，信号弱，海面特征不清晰。此外，现有的探测器难以在无人机的轻量化设计和探测精度之间取得平衡。为了解决这一问题，提出了一种新型的人船外检测变压器（mobo - detr）。一方面，引入Token增强层，对空间维度和通道维度进行细粒度过滤，减少后台查询带来的冗余编码；另一方面，提出了基于RepViT块的Effusion Fusion模块，通过解耦空间维度和通道维度之间的相互作用机制，有效消除了计算冗余。此外，为了填补现有人员落水检测基准数据集的空白，建立了ManOverboard基准数据集。在实验验证阶段，mobo - detr在ManOverboard和SeaDronesSeev2上进行。烧蚀实验表明，与基线相比，mobo - detr的轻量化性能提高了11.7%，APsmall提高了14.4%。在ManOverboard和SeaDronesSeev2上的对比实验验证了该方法的有效性，为人员落水检测提供了有效的解决方案。总的来说，本研究不仅推进了人员落水检测，而且显著提高了海上运输的弹性，最终保护了海员的生命，确保了世界重要贸易航线的可靠性。

{"title":"Maritime man-overboard search using a lightweight and efficient end-to-end detection transformer","authors":"Guokang Xu , Jianchuan Yin , Nini Wang , Zeguo Zhang","doi":"10.1016/j.jnlssr.2025.100267","DOIUrl":"10.1016/j.jnlssr.2025.100267","url":null,"abstract":"<div><div>Maritime transportation plays a crucial role in global economic trade. However, maritime accidents occur frequently, posing significant threats to the safety of seafarers. In search and rescue scenarios for man-overboard, unmanned aerial vehicles (UAVs) are gradually replacing manned aircraft and helicopters. Besides, detecting man-overboard is challenging because of their small pixel size, weak signals, and indistinct features on the ocean surface. Furthermore, existing detectors struggle to strike a balance between lightweight design for UAVs and detection accuracy. To address this issue, the novel Man-overboard Detection Transformer (MOB-DETR) is proposed. On the one hand, the Token Enhancement layer is introduced, which conducts fine-grained filtering of spatial and channel dimensions, reducing redundant encoding caused by background queries. On the other hand, the Effusion Fusion Module, based on the RepViT Block, is proposed, effectively eliminating computational redundancy by decoupling the interaction mechanisms between spatial and channel dimensions. Additionally, to fill the existing gap in benchmark datasets for detecting man-overboard, the ManOverboard benchmark dataset has been established. In the experimental validation phase, MOB-DETR is conducted on ManOverboard and SeaDronesSeev2. Ablation experiments show that MOB-DETR achieves 11.7 % better lightweight performance and 14.4 % higher <span><math><mrow><mi>A</mi><msub><mi>P</mi><mi>small</mi></msub></mrow></math></span> than baselines. Comparison experiments on ManOverboard and SeaDronesSeev2 validate its effectiveness, offering an efficient solution for man-overboard detection. Overall, this research not only advances man-overboard detection but also significantly enhances the resilience of maritime transportation, ultimately protecting seafarers' lives and ensuring the reliability of the world's essential trade routes.</div></div>","PeriodicalId":62710,"journal":{"name":"安全科学与韧性(英文)","volume":"7 2","pages":"Article 100267"},"PeriodicalIF":3.4,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Predicting models for critical velocity at the stair entrance under a two-story building fire 二层建筑火灾下楼梯入口临界速度预测模型

IF 3.4 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH

安全科学与韧性(英文)

Pub Date : 2025-09-27 DOI: 10.1016/j.jnlssr.2025.100266

Zeng Long , Cong Su , Chang Liu , Fazheng Chong , Fan Tong , Fengju Shang , Jiaqing Zhang , Jiansong Wu

The cross-floor spread of fire smoke in a two-story building hinders personal evacuation and leads to casualties. To address this issue, sufficient wind pressure needs to be applied at the stair to prevent smoke from entering the upper space, and the critical velocity is an important parameter worthy of attention. By conducting a series of model experiments and numerical simulations in this study, the critical velocity below the ceiling screen at the stair is investigated, considering the factors of heat release rate (HRR), fire source location, ceiling screen depth and number of stairs. The results show that the critical velocity is proportional to one-third of the HRR, and the proportional coefficient decreases as the fire source is farther from the stair and the ceiling screen depth increases. The prediction models for the critical velocity under fire sources located in front and back of the stairs are proposed, respectively, and the proposed model is validated under three stairs. Additionally, the proposed prediction model is compared with the velocity value required by the relevant standard, and some dangerous situations are identified that require further strengthened ventilation. This study can provide theoretical guidance for ventilation design and emergency response in actual engineering with similar structures.

在一栋两层楼的建筑中，火灾烟雾跨楼层蔓延，阻碍了人员疏散，并导致人员伤亡。为了解决这个问题，需要在楼梯处施加足够的风压，以防止烟雾进入上层空间，而临界风速是一个值得关注的重要参数。本研究通过一系列的模型实验和数值模拟，在考虑热释放率（HRR）、火源位置、顶棚深度和楼梯数量等因素的情况下，对楼梯顶棚下的临界速度进行了研究。结果表明：临界速度与HRR的1 / 3成正比，该比例系数随火源距离楼梯越远和顶棚遮挡深度的增加而减小；分别建立了台阶前后火源作用下的临界速度预测模型，并在三个台阶下对模型进行了验证。并将提出的预测模型与相关标准要求的流速值进行比较，识别出需要进一步加强通风的危险情况。该研究可为类似结构的实际工程通风设计和应急响应提供理论指导。

{"title":"Predicting models for critical velocity at the stair entrance under a two-story building fire","authors":"Zeng Long , Cong Su , Chang Liu , Fazheng Chong , Fan Tong , Fengju Shang , Jiaqing Zhang , Jiansong Wu","doi":"10.1016/j.jnlssr.2025.100266","DOIUrl":"10.1016/j.jnlssr.2025.100266","url":null,"abstract":"<div><div>The cross-floor spread of fire smoke in a two-story building hinders personal evacuation and leads to casualties. To address this issue, sufficient wind pressure needs to be applied at the stair to prevent smoke from entering the upper space, and the critical velocity is an important parameter worthy of attention. By conducting a series of model experiments and numerical simulations in this study, the critical velocity below the ceiling screen at the stair is investigated, considering the factors of heat release rate (HRR), fire source location, ceiling screen depth and number of stairs. The results show that the critical velocity is proportional to one-third of the HRR, and the proportional coefficient decreases as the fire source is farther from the stair and the ceiling screen depth increases. The prediction models for the critical velocity under fire sources located in front and back of the stairs are proposed, respectively, and the proposed model is validated under three stairs. Additionally, the proposed prediction model is compared with the velocity value required by the relevant standard, and some dangerous situations are identified that require further strengthened ventilation. This study can provide theoretical guidance for ventilation design and emergency response in actual engineering with similar structures.</div></div>","PeriodicalId":62710,"journal":{"name":"安全科学与韧性(英文)","volume":"7 2","pages":"Article 100266"},"PeriodicalIF":3.4,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145326926","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The impact of inaccurate supply-demand types for emergency supplies on the psychological pain of victims: Data from flood disasters in China 应急物资供需类型不准确对受害者心理痛苦的影响：来自中国洪涝灾害的数据

IF 3.4 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH

安全科学与韧性(英文)

Pub Date : 2025-09-11 DOI: 10.1016/j.jnlssr.2025.100265

Qian Yang, Feiyue Wang, Zihuan Wang, Bo Ma, Jiajie Lu, Leiwei Li

After a disaster, due to transportation constraints, transportation priority setting, emergency resource shortage and imprecise assessment of emergency resources, it is easy to cause an inaccurate match between the types of supply and demand for emergency resources. It aggravates the psychological pain of the victims. To address this gap, combining descriptive analysis, correlation analysis, and regression analysis, the mechanisms and internal pathways of imprecise emergency material supply-demand type on psychological pain were comprehensively examined. Based on Maslow's hierarchy of needs theory, four priority levels for emergency material types are proposed. A survey questionnaire was designed using the Numerical Rating Scale (NRS), which included 12 scenarios. Nine typical waterlogging sites in Changsha, Hunan Province, China, were selected as field survey sites, and 162 valid samples were collected through face-to-face interviews. The results show that: (1) Emergency material types have a significant impact on psychological pain, and this impact is related to the priority level of the material types. (2) The impact of the matching degree of emergency material supply-demand types on psychological pain is moderated by the priority level of the demanded material types, exhibiting a reverse compensation effect. (3) The impact of demographic factors on the degree of psychological pain experienced by victims shows complex differences. Most surprisingly, the higher the monthly income is, the more difficult it is to accept the inaccurate supply-demand of emergency supplies. Because high-income people have higher expectations for quality of life, their psychological pain is more obvious.

灾害发生后，由于交通运输的限制、运输优先级的设定、应急资源的短缺以及对应急资源评估的不精确，容易造成应急资源的供需类型匹配不准确。这加重了受害者的心理痛苦。为解决这一空白，结合描述性分析、相关性分析和回归分析，全面考察了不精确应急物资供需类型对心理疼痛的影响机制和内在途径。基于马斯洛需求层次理论，提出了应急物资类型的四个优先级。采用数值评定量表（NRS）设计调查问卷，共设12个场景。选取中国湖南省长沙市9个典型内涝点作为实地调查点，通过面对面访谈收集有效样本162份。结果表明：(1)应急物资类型对心理疼痛有显著影响，且这种影响与物资类型的优先级有关。(2)应急物资供需类型匹配程度对心理痛苦的影响受需求物资类型优先级的调节，呈现反向补偿效应。(3)人口学因素对受害者心理痛苦程度的影响存在复杂差异。最令人惊讶的是，月收入越高，就越难以接受不准确的应急物资供需。因为高收入人群对生活质量的期望更高，他们的心理痛苦也更明显。

{"title":"The impact of inaccurate supply-demand types for emergency supplies on the psychological pain of victims: Data from flood disasters in China","authors":"Qian Yang, Feiyue Wang, Zihuan Wang, Bo Ma, Jiajie Lu, Leiwei Li","doi":"10.1016/j.jnlssr.2025.100265","DOIUrl":"10.1016/j.jnlssr.2025.100265","url":null,"abstract":"<div><div>After a disaster, due to transportation constraints, transportation priority setting, emergency resource shortage and imprecise assessment of emergency resources, it is easy to cause an inaccurate match between the types of supply and demand for emergency resources. It aggravates the psychological pain of the victims. To address this gap, combining descriptive analysis, correlation analysis, and regression analysis, the mechanisms and internal pathways of imprecise emergency material supply-demand type on psychological pain were comprehensively examined. Based on Maslow's hierarchy of needs theory, four priority levels for emergency material types are proposed. A survey questionnaire was designed using the Numerical Rating Scale (NRS), which included 12 scenarios. Nine typical waterlogging sites in Changsha, Hunan Province, China, were selected as field survey sites, and 162 valid samples were collected through face-to-face interviews. The results show that: (1) Emergency material types have a significant impact on psychological pain, and this impact is related to the priority level of the material types. (2) The impact of the matching degree of emergency material supply-demand types on psychological pain is moderated by the priority level of the demanded material types, exhibiting a reverse compensation effect. (3) The impact of demographic factors on the degree of psychological pain experienced by victims shows complex differences. Most surprisingly, the higher the monthly income is, the more difficult it is to accept the inaccurate supply-demand of emergency supplies. Because high-income people have higher expectations for quality of life, their psychological pain is more obvious.</div></div>","PeriodicalId":62710,"journal":{"name":"安全科学与韧性(英文)","volume":"7 2","pages":"Article 100265"},"PeriodicalIF":3.4,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145789664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Fusing emergency data and seasonal effects for urban rail transit safety assessment: A severity probability distribution model 融合应急数据和季节影响的城市轨道交通安全评估：一个严重性概率分布模型

IF 3.4 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH

安全科学与韧性(英文)

Pub Date : 2025-09-10 DOI: 10.1016/j.jnlssr.2025.100264

Bosong Fan , Dongyu Luo , Chunfu Shao

An analysis of Beijing urban rail transit emergency data from 2016 to 2018 is conducted to identify the factors influencing emergency severity. A severity probability distribution model is proposed, incorporating seasonal effects, multi-dimensional features (passenger, vehicle, line, environment), random parameters, interaction terms, and weather variables (temperature, humidity, wind speed) to capture severity probability changes driven by seasonally influenced factors. High performance is achieved by the model, with 92.10 % accuracy, 95.17 % sensitivity, and 87.58 % specificity. Five key factors—bad weather (correlation = -0.0611, p < 0.001), track failure (correlation = -0.0503, p < 0.001), signal failure (correlation = -0.0483, p < 0.001), urban area (correlation = -0.0441, p < 0.001), and vehicle failure (correlation = -0.0435, p < 0.001)—are identified, with their interactions amplifying risks, particularly the coupling effect between bad weather and train system failures. Notable seasonal effects are observed: In summer, high temperature and humidity increase vehicle and equipment failures; in autumn, temperature fluctuations raise signaling anomaly risks; in winter, low temperatures elevate track and equipment failure risks; and in spring and winter, windy weather exacerbates risks for suburban elevated lines. This study highlights the need for refined risk identification and response mechanisms at the seasonal and regional levels, such as targeted inspections and weather-triggered controls, to enhance operational safety in urban rail transit under complex weather conditions.

对2016 - 2018年北京市城市轨道交通突发事件数据进行分析，找出影响突发事件严重程度的因素。提出了一种综合季节效应、多维特征（乘客、车辆、线路、环境）、随机参数、交互项和天气变量（温度、湿度、风速）的严重性概率分布模型，以捕捉季节影响因素驱动下的严重性概率变化。该模型的准确率为92.10%，灵敏度为95.17%，特异性为87.58%。确定了五个关键因素——恶劣天气（相关性= -0.0611,p < 0.001）、轨道故障（相关性= -0.0503,p < 0.001）、信号故障（相关性= -0.0483,p < 0.001）、城区（相关性= -0.0441,p < 0.001）和车辆故障（相关性= -0.0435,p < 0.001），它们之间的相互作用放大了风险，特别是恶劣天气和列车系统故障之间的耦合效应。季节效应显著：夏季高温高湿使车辆和设备故障增加；秋季气温波动增加信号异常风险；冬季，低温增加了轨道和设备故障的风险；在春季和冬季，多风的天气加剧了郊区高架线路的风险。本研究强调需要在季节和区域层面完善风险识别和响应机制，例如有针对性的检查和天气触发的控制，以提高复杂天气条件下城市轨道交通的运营安全。

{"title":"Fusing emergency data and seasonal effects for urban rail transit safety assessment: A severity probability distribution model","authors":"Bosong Fan , Dongyu Luo , Chunfu Shao","doi":"10.1016/j.jnlssr.2025.100264","DOIUrl":"10.1016/j.jnlssr.2025.100264","url":null,"abstract":"<div><div>An analysis of Beijing urban rail transit emergency data from 2016 to 2018 is conducted to identify the factors influencing emergency severity. A severity probability distribution model is proposed, incorporating seasonal effects, multi-dimensional features (passenger, vehicle, line, environment), random parameters, interaction terms, and weather variables (temperature, humidity, wind speed) to capture severity probability changes driven by seasonally influenced factors. High performance is achieved by the model, with 92.10 % accuracy, 95.17 % sensitivity, and 87.58 % specificity. Five key factors—bad weather (correlation = -0.0611, <em>p</em> < 0.001), track failure (correlation = -0.0503, <em>p</em> < 0.001), signal failure (correlation = -0.0483, <em>p</em> < 0.001), urban area (correlation = -0.0441, <em>p</em> < 0.001), and vehicle failure (correlation = -0.0435, <em>p</em> < 0.001)—are identified, with their interactions amplifying risks, particularly the coupling effect between bad weather and train system failures. Notable seasonal effects are observed: In summer, high temperature and humidity increase vehicle and equipment failures; in autumn, temperature fluctuations raise signaling anomaly risks; in winter, low temperatures elevate track and equipment failure risks; and in spring and winter, windy weather exacerbates risks for suburban elevated lines. This study highlights the need for refined risk identification and response mechanisms at the seasonal and regional levels, such as targeted inspections and weather-triggered controls, to enhance operational safety in urban rail transit under complex weather conditions.</div></div>","PeriodicalId":62710,"journal":{"name":"安全科学与韧性(英文)","volume":"7 2","pages":"Article 100264"},"PeriodicalIF":3.4,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Intelligent fire modeling in wildland-urban interface: A comprehensive review of current progress, challenges, and future perspectives 荒地-城市界面中的智能火灾建模：当前进展、挑战和未来展望的综合回顾

IF 3.4 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH

安全科学与韧性(英文)

Pub Date : 2025-09-06 DOI: 10.1016/j.jnlssr.2025.100252

Ao Zheng , Rui Ba , Wenyu Jiang , Zijun Chen , Menghao He , Yuansheng Hua , Song Zhu , Jiasong Zhu , Guochao Liu , Zhuojie Zhu , Xinyi Han

Wildfire prevention and control, especially within complex Wildland-Urban Interface (WUI), face escalating challenges due to the synergistic impacts of climate change and expanding urban frontiers. While advanced wildfire spread prediction models are essential for developing disaster-resilient emergency systems, the increasing complexities of WUI wildfire scenarios highlight critical limitations in current modeling approaches. These complexities include the dynamic interactions between vegetation and built environments, the demands of multi-scale spatiotemporal forecasting, and the challenges of cross-platform integration. To better understand and address these challenges, this paper establishes a novel tripartite analytical framework through a systematic review of peer-reviewed studies from the Scopus and Web of Science databases: (1) fire combustion characterization, (2) fire dynamics mechanism, and (3) fire management system. Our critical analysis identifies three persistent research challenges, including modeling fire behaviors in heterogeneous WUI scenarios; balancing fire dynamics with computational speed, accuracy, and resolution; and managing and applying models in highly integrated systems. This study concludes with actionable priorities for subsequent research, providing methodological guidelines for model developers and evidence-based integration pathways for emergency management systems, particularly in addressing critical infrastructure protection in rapidly urbanizing, fire-prone regions.

由于气候变化和城市边界扩大的协同影响，野火防控，特别是复杂的野火-城市界面（WUI）的野火防控面临着日益严峻的挑战。虽然先进的野火蔓延预测模型对于开发抗灾应急系统至关重要，但WUI野火场景的日益复杂性突出了当前建模方法的关键局限性。这些复杂性包括植被与建筑环境之间的动态相互作用、多尺度时空预测的需求以及跨平台集成的挑战。为了更好地理解和应对这些挑战，本文通过系统回顾来自Scopus和Web of Science数据库的同行评议研究，建立了一个新的三方分析框架：(1)火灾燃烧特性，(2)火灾动力学机制，(3)火灾管理系统。我们的批判性分析确定了三个持续存在的研究挑战，包括在异构WUI场景中模拟火灾行为；平衡火动力学与计算速度，准确性和分辨率；以及在高度集成的系统中管理和应用模型。本研究总结了后续研究的可操作优先事项，为模型开发者提供了方法指南，并为应急管理系统提供了基于证据的整合途径，特别是在快速城市化、火灾易发地区解决关键基础设施保护问题。

{"title":"Intelligent fire modeling in wildland-urban interface: A comprehensive review of current progress, challenges, and future perspectives","authors":"Ao Zheng , Rui Ba , Wenyu Jiang , Zijun Chen , Menghao He , Yuansheng Hua , Song Zhu , Jiasong Zhu , Guochao Liu , Zhuojie Zhu , Xinyi Han","doi":"10.1016/j.jnlssr.2025.100252","DOIUrl":"10.1016/j.jnlssr.2025.100252","url":null,"abstract":"<div><div>Wildfire prevention and control, especially within complex Wildland-Urban Interface (WUI), face escalating challenges due to the synergistic impacts of climate change and expanding urban frontiers. While advanced wildfire spread prediction models are essential for developing disaster-resilient emergency systems, the increasing complexities of WUI wildfire scenarios highlight critical limitations in current modeling approaches. These complexities include the dynamic interactions between vegetation and built environments, the demands of multi-scale spatiotemporal forecasting, and the challenges of cross-platform integration. To better understand and address these challenges, this paper establishes a novel tripartite analytical framework through a systematic review of peer-reviewed studies from the Scopus and Web of Science databases: (1) fire combustion characterization, (2) fire dynamics mechanism, and (3) fire management system. Our critical analysis identifies three persistent research challenges, including modeling fire behaviors in heterogeneous WUI scenarios; balancing fire dynamics with computational speed, accuracy, and resolution; and managing and applying models in highly integrated systems. This study concludes with actionable priorities for subsequent research, providing methodological guidelines for model developers and evidence-based integration pathways for emergency management systems, particularly in addressing critical infrastructure protection in rapidly urbanizing, fire-prone regions.</div></div>","PeriodicalId":62710,"journal":{"name":"安全科学与韧性(英文)","volume":"7 2","pages":"Article 100252"},"PeriodicalIF":3.4,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145789662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Study on combustion characteristics of cables with spacing arrangement under different external radiation conditions 不同外辐射条件下布置间距电缆燃烧特性研究

IF 3.4 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH

安全科学与韧性(英文)

Pub Date : 2025-09-06 DOI: 10.1016/j.jnlssr.2025.100255

Tong Xu , Xiaole Dai , Kai Wang , Fazheng Chong , Fengju Shang , Chang Liu

A series of experiments was conducted to explore the influence of cable spacing on combustion characteristics in real cable installation, considering various external radiation intensities (30, 50 kW/m²) and cable spacings (0, 2.5, 5 cm). A comparative analysis of the combustion characteristics (such as heat release rate (HRR), combustion gas, and mass loss) was conducted, and the HRR calculation for cables that were not fully filled in the sample tray was revised as well. It could be found that with the increasing radiation intensities, the peak concentrations of CO and CO₂ increased (O₂ concentration decreased), and the interval between the two peaks shortened. The HRR curves of cables with different spacings all presented two peaks under two radiation intensities. The max-peak HRR occurs at D_d = 2.5 cm, and the double max-peak HRRs are 582 kW/m² and 407 kW/m² under radiation intensities of 50 kW/m² and 35 kW/m², respectively. This is because when D_d = 0 cm, the thermal feedback effect between cables is relatively enhanced, while the air entrainment between the cables is weakened. When D_d = 5 cm, the thermal feedback effect is weakened. When D_d = 2.5 cm, both the air entrainment and the thermal feedback are strengthened, and the peak HRR occurs. The above results could provide data to support fire safety design and the emergency response to cable laying.

在实际电缆安装中，考虑不同的外辐射强度（30、50 kW/m2）和电缆间距（0、2.5、5 cm），进行了一系列实验，探讨了电缆间距对燃烧特性的影响。对比分析了燃烧特性（热释放率（HRR）、燃烧气体、质量损失等），并修正了样品盘内未完全填充电缆的HRR计算。可以发现，随着辐射强度的增加，CO和CO₂的峰值浓度增加（O₂浓度降低），两峰之间的间隔缩短。不同间距电缆的HRR曲线在两种辐射强度下均呈现两个峰值。辐射强度为50 kW/m2和35 kW/m2时，双峰HRR分别为582 kW/m2和407 kW/m2。这是因为当Dd = 0 cm时，电缆间的热反馈效应相对增强，而电缆间的夹带气流减弱。当Dd = 5 cm时，热反馈效应减弱。当Dd = 2.5 cm时，夹带空气和热反馈都增强，HRR出现峰值。上述结果可为电缆敷设的消防安全设计和应急响应提供数据支持。

{"title":"Study on combustion characteristics of cables with spacing arrangement under different external radiation conditions","authors":"Tong Xu , Xiaole Dai , Kai Wang , Fazheng Chong , Fengju Shang , Chang Liu","doi":"10.1016/j.jnlssr.2025.100255","DOIUrl":"10.1016/j.jnlssr.2025.100255","url":null,"abstract":"<div><div>A series of experiments was conducted to explore the influence of cable spacing on combustion characteristics in real cable installation, considering various external radiation intensities (30, 50 kW/m<sup>2</sup>) and cable spacings (0, 2.5, 5 cm). A comparative analysis of the combustion characteristics (such as heat release rate (HRR), combustion gas, and mass loss) was conducted, and the HRR calculation for cables that were not fully filled in the sample tray was revised as well. It could be found that with the increasing radiation intensities, the peak concentrations of CO and CO₂ increased (O₂ concentration decreased), and the interval between the two peaks shortened. The HRR curves of cables with different spacings all presented two peaks under two radiation intensities. The max-peak HRR occurs at <em>D<sub>d</sub></em> = 2.5 cm, and the double max-peak HRRs are 582 kW/m<sup>2</sup> and 407 kW/m<sup>2</sup> under radiation intensities of 50 kW/m<sup>2</sup> and 35 kW/m<sup>2</sup>, respectively. This is because when <em>D<sub>d</sub></em> = 0 cm, the thermal feedback effect between cables is relatively enhanced, while the air entrainment between the cables is weakened. When <em>D<sub>d</sub></em> = 5 cm, the thermal feedback effect is weakened. When <em>D<sub>d</sub></em> = 2.5 cm, both the air entrainment and the thermal feedback are strengthened, and the peak HRR occurs. The above results could provide data to support fire safety design and the emergency response to cable laying.</div></div>","PeriodicalId":62710,"journal":{"name":"安全科学与韧性(英文)","volume":"7 2","pages":"Article 100255"},"PeriodicalIF":3.4,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145789663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A lightweight four-channel multi-modal model to improve computational performance of automated fire detection 一个轻量级的四通道多模态模型，以提高自动火灾探测的计算性能

IF 3.4 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH

安全科学与韧性(英文)

Pub Date : 2025-09-05 DOI: 10.1016/j.jnlssr.2025.100254

Jingshuo Yu , Qian Chen

The urgent need for advanced fire detection methods stems from the increased intensity of fire incidents, which cause massive property loss and irreversible damage. To overcome the limitations of traditional fire detection methods, such as those of smoke detectors, fire detection based on computer vision (CV) algorithms has been adopted to improve detection accuracy. Compared to single-modal fire detection, multi-modal fire detection has gained attention because it leverages the richer information present in both RGB and thermal images. However, prevalent multi-modal fire detection methods significantly increase model complexity by requiring two separate streams in the backbone to process RGB and thermal images independently. To address this issue, this paper proposes a four-channel single-stream fire detection method based on YOLOv5, which concatenates RGB and thermal images to form the required four-channel input. Comparison experiments with dual-stream YOLOv5 models using add fusion and transformer fusion demonstrate that the four-channel single-stream model reduces model complexity while improving detection accuracy. To further enhance detection accuracy and reduce model complexity, this study redesigned YOLOv5’s C3 module by integrating the convolutional block attention module (CBAM) to form the C3CBAM module and introduced the SCYLLA-Intersection over Union (SIoU) loss function. By comparing its performance with that of state-of-the-art (SOTA) models in multi-modal object detection, such as the YOLOv5-based dual-stream model, this study shows that the proposed approach improves detection in the diverse conditions presented in the selected dataset.

由于火灾事故的频繁发生，造成了巨大的财产损失和不可逆转的损失，迫切需要先进的火灾探测方法。为了克服传统火灾探测方法（如烟雾探测器）的局限性，采用基于计算机视觉（CV）算法的火灾探测来提高探测精度。与单模态火灾探测相比，多模态火灾探测受到了人们的关注，因为它利用了RGB和热图像中更丰富的信息。然而，流行的多模态火灾探测方法需要在主干中两个独立的流来独立处理RGB和热图像，从而显著增加了模型的复杂性。针对这一问题，本文提出了一种基于YOLOv5的四通道单流火灾检测方法，该方法将RGB图像和热图像拼接成所需的四通道输入。采用添加融合和变压器融合的双流YOLOv5模型对比实验表明，四通道单流模型在降低模型复杂度的同时提高了检测精度。为了进一步提高检测精度，降低模型复杂度，本研究对YOLOv5的C3模块进行了重新设计，将卷积块注意模块（CBAM）集成为C3CBAM模块，并引入了SCYLLA-Intersection over Union （SIoU）损失函数。通过将该方法与基于yolov5的双流模型等最先进的SOTA模型在多模态目标检测中的性能进行比较，本研究表明，该方法可以提高所选数据集中不同条件下的检测效果。

{"title":"A lightweight four-channel multi-modal model to improve computational performance of automated fire detection","authors":"Jingshuo Yu , Qian Chen","doi":"10.1016/j.jnlssr.2025.100254","DOIUrl":"10.1016/j.jnlssr.2025.100254","url":null,"abstract":"<div><div>The urgent need for advanced fire detection methods stems from the increased intensity of fire incidents, which cause massive property loss and irreversible damage. To overcome the limitations of traditional fire detection methods, such as those of smoke detectors, fire detection based on computer vision (CV) algorithms has been adopted to improve detection accuracy. Compared to single-modal fire detection, multi-modal fire detection has gained attention because it leverages the richer information present in both RGB and thermal images. However, prevalent multi-modal fire detection methods significantly increase model complexity by requiring two separate streams in the backbone to process RGB and thermal images independently. To address this issue, this paper proposes a four-channel single-stream fire detection method based on YOLOv5, which concatenates RGB and thermal images to form the required four-channel input. Comparison experiments with dual-stream YOLOv5 models using add fusion and transformer fusion demonstrate that the four-channel single-stream model reduces model complexity while improving detection accuracy. To further enhance detection accuracy and reduce model complexity, this study redesigned YOLOv5’s C3 module by integrating the convolutional block attention module (CBAM) to form the C3CBAM module and introduced the SCYLLA-Intersection over Union (SIoU) loss function. By comparing its performance with that of state-of-the-art (SOTA) models in multi-modal object detection, such as the YOLOv5-based dual-stream model, this study shows that the proposed approach improves detection in the diverse conditions presented in the selected dataset.</div></div>","PeriodicalId":62710,"journal":{"name":"安全科学与韧性(英文)","volume":"7 2","pages":"Article 100254"},"PeriodicalIF":3.4,"publicationDate":"2025-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145736496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Construction and application of knowledge graph for urban agglomeration emergency rescue collaboration in earthquake disasters 城市群地震灾害应急救援协同知识图谱的构建与应用

IF 3.4 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH

安全科学与韧性(英文)

Pub Date : 2025-09-04 DOI: 10.1016/j.jnlssr.2025.100253

Lin Zhang , Qichen Wang , Yanjun Guo , Xiangliang Tian , Lin Qi

This study proposes an emergency rescue collaboration knowledge graph construction method for urban agglomeration in earthquake disasters. Based on the collection of 22 earthquake disaster emergency plans published on the official websites of multiple cities in the Chengdu-Chongqing urban agglomeration in China, earthquake disaster emergency rescue data from the Red Cross Society of Sichuan Province and Chongqing City, and historical rescue information from the China Blue Sky rescue team, this study defines and extracts six types of entities including rescue entities, policy documents, rescue actions, rescue information, rescue supplies, and emergency response levels. A knowledge graph pattern layer is established using a hybrid approach of top-down and bottom-up, including concept layer, relationship layer, rule layer, and instance layer. This study extracts earthquake disaster emergency rescue collaboration knowledge information from collected data sources, and YEDDA software is used for knowledge fusion, thus constructing a knowledge graph data layer. The data is stored in the Neo4j graph database as triplets (entity-relation-entity). Visual representation and retrieval are used to achieve the query, association, and inference of emergency rescue collaboration information for urban agglomeration in earthquake disasters. The 2022 Luding earthquake disaster in Ganzi Tibetan Autonomous Prefecture, China is selected as a typical case, and verified the effectiveness and reliability by inputting the case into the emergency rescue collaboration knowledge graph which was constructed in this study. The results indicate that the constructed knowledge graph provides intelligent decision support for earthquake disaster emergency rescue collaboration in urban agglomeration, effectively improves the performance of earthquake disaster emergency rescue, and provides new ideas and methods for earthquake disaster rescue and reduction.

本文提出了一种城市群地震灾害应急救援协同知识图谱构建方法。本研究在收集成渝城市群多个城市官方网站发布的22份地震灾害应急预案、四川省红十字会和重庆市红十字会的地震灾害应急救援数据、中国蓝天救援队的历史救援信息的基础上，定义并提取了救援主体、政策文件、救援行动、救援信息、救援物资和应急响应水平。采用自顶向下和自底向上混合的方法建立知识图谱模式层，包括概念层、关系层、规则层和实例层。本研究从采集到的数据源中提取地震灾害应急救援协同知识信息，利用YEDDA软件进行知识融合，构建知识图谱数据层。数据以三元组（实体-关系-实体）的形式存储在Neo4j图形数据库中。采用可视化表示和检索技术，实现了地震灾害下城市群应急救援协同信息的查询、关联和推理。结果表明，构建的知识图谱为城市群地震灾害应急救援协同提供了智能决策支持，有效提高了地震灾害应急救援性能，为地震灾害救援减灾提供了新的思路和方法。

{"title":"Construction and application of knowledge graph for urban agglomeration emergency rescue collaboration in earthquake disasters","authors":"Lin Zhang , Qichen Wang , Yanjun Guo , Xiangliang Tian , Lin Qi","doi":"10.1016/j.jnlssr.2025.100253","DOIUrl":"10.1016/j.jnlssr.2025.100253","url":null,"abstract":"<div><div>This study proposes an emergency rescue collaboration knowledge graph construction method for urban agglomeration in earthquake disasters. Based on the collection of 22 earthquake disaster emergency plans published on the official websites of multiple cities in the Chengdu-Chongqing urban agglomeration in China, earthquake disaster emergency rescue data from the Red Cross Society of Sichuan Province and Chongqing City, and historical rescue information from the China Blue Sky rescue team, this study defines and extracts six types of entities including rescue entities, policy documents, rescue actions, rescue information, rescue supplies, and emergency response levels. A knowledge graph pattern layer is established using a hybrid approach of top-down and bottom-up, including concept layer, relationship layer, rule layer, and instance layer. This study extracts earthquake disaster emergency rescue collaboration knowledge information from collected data sources, and YEDDA software is used for knowledge fusion, thus constructing a knowledge graph data layer. The data is stored in the Neo4j graph database as triplets (entity-relation-entity). Visual representation and retrieval are used to achieve the query, association, and inference of emergency rescue collaboration information for urban agglomeration in earthquake disasters. The 2022 Luding earthquake disaster in Ganzi Tibetan Autonomous Prefecture, China is selected as a typical case, and verified the effectiveness and reliability by inputting the case into the emergency rescue collaboration knowledge graph which was constructed in this study. The results indicate that the constructed knowledge graph provides intelligent decision support for earthquake disaster emergency rescue collaboration in urban agglomeration, effectively improves the performance of earthquake disaster emergency rescue, and provides new ideas and methods for earthquake disaster rescue and reduction.</div></div>","PeriodicalId":62710,"journal":{"name":"安全科学与韧性(英文)","volume":"7 1","pages":"Article 100253"},"PeriodicalIF":3.4,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145362247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SIL assessment of safety instrumented systems in oil and gas stations based on STPA-Bow-tie 基于STPA-Bow-tie的油气站安全仪表系统SIL评价

IF 3.4 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH

安全科学与韧性(英文)

Pub Date : 2025-09-02 DOI: 10.1016/j.jnlssr.2025.100251

Wencong Ye , Jinjiang Wang , Zhenqiang Wei , Zheng Wang , Laibin Zhang

The growing imperative for safe and intelligent operation of long-distance oil and gas pipeline systems has led to increased deployment of interlock systems at processing stations, but their reliability and compliance are difficult to guarantee. To achieve process hazard analysis (PHA) and safety integrity level (SIL) assessment in complex hazardous scenarios of oil and gas stations, this paper presents a safety instrumented system (SIS) SIL assessment method based on a combination of system theory, process analysis and the Bow-tie model (STPA-Bow-tie). First, the boundary of the station process system is determined, and unsafe control actions (UCAs) and their key causes are identified based on the hierarchical control structure model of the oil and gas stations. Then, the element mapping relationship among STPA, layers of protection analysis (LOPA), and Bow-tie is proposed, and the Bow-tie model based on Simulink is constructed to realize LOPA, so as to quantify the required risk reduction factor (RRF) and determine the target SIL level of the safety instrument function (SIF). Finally, a Markov model verifies the safety integrity of the interlock circuit. Taking the SIL assessment of an SIS system in a certain oil transfer station as an example, the proposed method demonstrates equivalent SIL determination accuracy to the traditional HAZOP-LOPA method while providing 24 % higher analytical precision and superior visualization of multi-cause/consequence coupling in complex hazard scenarios.

随着长输油气管道系统安全、智能化运行的日益迫切，处理站的联锁系统部署越来越多，但联锁系统的可靠性和合规性难以保证。为实现油气田复杂危险场景下的过程危害分析（PHA）和安全完整性等级（SIL）评价，提出了一种基于系统论、过程分析和蝴蝶结模型（STPA-Bow-tie）相结合的安全仪表系统SIL评价方法。首先，确定加油站过程系统边界，基于油气站分层控制结构模型，识别不安全控制动作及其关键原因；然后，提出了STPA、防护层分析（LOPA）和Bow-tie之间的元素映射关系，并构建了基于Simulink的Bow-tie模型来实现LOPA，从而量化所需的风险降低因子（RRF），确定安全仪表功能（SIF）的目标SIL水平。最后通过马尔可夫模型验证了联锁电路的安全完整性。以某石油中转站SIS系统的SIL评估为例，该方法的SIL测定精度与传统HAZOP-LOPA方法相当，分析精度提高24%，在复杂危害情景下多因/后果耦合的可视化效果优于传统HAZOP-LOPA方法。

{"title":"SIL assessment of safety instrumented systems in oil and gas stations based on STPA-Bow-tie","authors":"Wencong Ye , Jinjiang Wang , Zhenqiang Wei , Zheng Wang , Laibin Zhang","doi":"10.1016/j.jnlssr.2025.100251","DOIUrl":"10.1016/j.jnlssr.2025.100251","url":null,"abstract":"<div><div>The growing imperative for safe and intelligent operation of long-distance oil and gas pipeline systems has led to increased deployment of interlock systems at processing stations, but their reliability and compliance are difficult to guarantee. To achieve process hazard analysis (PHA) and safety integrity level (SIL) assessment in complex hazardous scenarios of oil and gas stations, this paper presents a safety instrumented system (SIS) SIL assessment method based on a combination of system theory, process analysis and the Bow-tie model (STPA-Bow-tie). First, the boundary of the station process system is determined, and unsafe control actions (UCAs) and their key causes are identified based on the hierarchical control structure model of the oil and gas stations. Then, the element mapping relationship among STPA, layers of protection analysis (LOPA), and Bow-tie is proposed, and the Bow-tie model based on Simulink is constructed to realize LOPA, so as to quantify the required risk reduction factor (RRF) and determine the target SIL level of the safety instrument function (SIF). Finally, a Markov model verifies the safety integrity of the interlock circuit. Taking the SIL assessment of an SIS system in a certain oil transfer station as an example, the proposed method demonstrates equivalent SIL determination accuracy to the traditional HAZOP-LOPA method while providing 24 % higher analytical precision and superior visualization of multi-cause/consequence coupling in complex hazard scenarios.</div></div>","PeriodicalId":62710,"journal":{"name":"安全科学与韧性(英文)","volume":"7 2","pages":"Article 100251"},"PeriodicalIF":3.4,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145520269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Machine learning for occupational accident analysis: Applications, challenges, and future directions 机器学习用于职业事故分析：应用、挑战和未来方向

IF 3.4 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH

安全科学与韧性(英文)

Pub Date : 2025-08-29 DOI: 10.1016/j.jnlssr.2025.100250

Izuchukwu Chukwuma Obasi, Pericles Cheng, Cleo Varianou-Mikellidou, Christos Dimopoulos, Georgios Boustras

Machine learning (ML) drives progress in occupational accident prevention across diverse sectors. However, significant challenges persist in aligning these tools with practical safety needs, including accurate risk assessment, incident prediction, and targeted prevention strategies. While prior reviews focused narrowly on specific industries or data types, this study presents a comprehensive analysis of ML models in accident analysis, categorizing them by accident type, industry application, and modeling methodology. This study addresses critical challenges in ML model development—such as data quality, hyperparameter tuning, and managing class imbalances—and examines less-discussed topics, including explanatory variable selection and strategies for mitigating overfitting. This review thoroughly assesses the current state of ML-based accident prediction, highlighting critical gaps, methodological limitations, and potential research directions. By analyzing 504 studies across three perspectives—Accident Type, Industry Application, and Modeling Methodology—this review identifies pressing challenges, including (1) limitations in data quality and availability, especially for real-time sources; (2) inadequate model interpretability across applications; (3) difficulties in handling imbalanced accident datasets; and (4) the lack of an integrated framework for incorporating proactive data and industry-specific risk factors. The findings outline a roadmap for advancing ML in occupational safety by enhancing model robustness, improving interpretability, and expanding data sources. This review aims to better align ML applications with safety objectives, promoting data-driven approaches for effective accident analysis and prevention across industries.

机器学习（ML）推动了各个部门在职业事故预防方面的进步。然而，在将这些工具与实际安全需求相结合方面仍然存在重大挑战，包括准确的风险评估、事件预测和有针对性的预防策略。虽然之前的评论只关注特定的行业或数据类型，但本研究对事故分析中的ML模型进行了全面分析，并根据事故类型、行业应用和建模方法对其进行了分类。本研究解决了机器学习模型开发中的关键挑战，如数据质量、超参数调优和管理类不平衡，并研究了较少讨论的主题，包括解释变量选择和减轻过拟合的策略。本文全面评估了基于机器学习的事故预测的现状，强调了关键差距、方法局限性和潜在的研究方向。通过从事故类型、行业应用和建模方法三个角度分析504项研究，本综述确定了紧迫的挑战，包括：(1)数据质量和可用性的限制，特别是对于实时来源；(2)不同应用的模型可解释性不足；(3)不平衡事故数据集处理困难；(4)缺乏整合前瞻性数据和行业特定风险因素的综合框架。研究结果概述了通过增强模型鲁棒性、提高可解释性和扩展数据源来推进ML在职业安全中的发展路线图。此次审查旨在更好地将机器学习应用与安全目标结合起来，促进数据驱动的方法，以便在各个行业进行有效的事故分析和预防。

{"title":"Machine learning for occupational accident analysis: Applications, challenges, and future directions","authors":"Izuchukwu Chukwuma Obasi, Pericles Cheng, Cleo Varianou-Mikellidou, Christos Dimopoulos, Georgios Boustras","doi":"10.1016/j.jnlssr.2025.100250","DOIUrl":"10.1016/j.jnlssr.2025.100250","url":null,"abstract":"<div><div>Machine learning (ML) drives progress in occupational accident prevention across diverse sectors. However, significant challenges persist in aligning these tools with practical safety needs, including accurate risk assessment, incident prediction, and targeted prevention strategies. While prior reviews focused narrowly on specific industries or data types, this study presents a comprehensive analysis of ML models in accident analysis, categorizing them by accident type, industry application, and modeling methodology. This study addresses critical challenges in ML model development—such as data quality, hyperparameter tuning, and managing class imbalances—and examines less-discussed topics, including explanatory variable selection and strategies for mitigating overfitting. This review thoroughly assesses the current state of ML-based accident prediction, highlighting critical gaps, methodological limitations, and potential research directions. By analyzing 504 studies across three perspectives—Accident Type, Industry Application, and Modeling Methodology—this review identifies pressing challenges, including (1) limitations in data quality and availability, especially for real-time sources; (2) inadequate model interpretability across applications; (3) difficulties in handling imbalanced accident datasets; and (4) the lack of an integrated framework for incorporating proactive data and industry-specific risk factors. The findings outline a roadmap for advancing ML in occupational safety by enhancing model robustness, improving interpretability, and expanding data sources. This review aims to better align ML applications with safety objectives, promoting data-driven approaches for effective accident analysis and prevention across industries.</div></div>","PeriodicalId":62710,"journal":{"name":"安全科学与韧性(英文)","volume":"7 1","pages":"Article 100250"},"PeriodicalIF":3.4,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145048450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

安全科学与韧性(英文)最新文献