Safety SciencePub Date : 2025-02-11DOI: 10.1016/j.ssci.2025.106814
Haonan Qi , Zhipeng Zhou , Patrick Manu , Nan Li
{"title":"Falling risk analysis at workplaces through an accident data-driven approach based upon hybrid artificial intelligence (AI) techniques","authors":"Haonan Qi , Zhipeng Zhou , Patrick Manu , Nan Li","doi":"10.1016/j.ssci.2025.106814","DOIUrl":"10.1016/j.ssci.2025.106814","url":null,"abstract":"<div><div>This study proposed an accident data-driven approach using hybrid AI techniques for the quantification of falling risks at workplaces. Six machine learning models and one ensemble learning model were deployed for automatic extraction of causal factors. These causal factors were taken as main nodes in the falling risk Bayesian network (FRBN). Data-driven and knowledge-driven methods were combined for structure learning of the FRBN, based upon algorithms of hill climbing and tree augmented naive Bayes firstly and modification of FRBN through incorporation of knowledge. Sensitive causal factors were determined using parameter-based and evidence-based sensitivity analysis approaches. The FRBN was further adopted for forward and backward causal inferences. The accident data-driven approach through hybrid AI techniques contributes to substantial learning from fall-related accidents. Measures would be tailored according to causal inferences within the FRBN, so that the probability of falling risk will be reduced and negative impacts of fall-from-height (FFH) accidents will be decreased.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"185 ","pages":"Article 106814"},"PeriodicalIF":4.7,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143387314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Safety SciencePub Date : 2025-02-10DOI: 10.1016/j.ssci.2025.106806
Xiao Zhao , Ze-Rui Xiang , Zutao Zhang , Tie-Cheng Ding , Hai-Nan Liu , Hong-Bao Wang , Rui Zou , Yu Wang
{"title":"Factors affecting the visual ergonomics of train drivers in VR simulation driving: Snow and Ice line environment and train speed","authors":"Xiao Zhao , Ze-Rui Xiang , Zutao Zhang , Tie-Cheng Ding , Hai-Nan Liu , Hong-Bao Wang , Rui Zou , Yu Wang","doi":"10.1016/j.ssci.2025.106806","DOIUrl":"10.1016/j.ssci.2025.106806","url":null,"abstract":"<div><div>Optimal visual ergonomics is a crucial cornerstone for the safe operation of the railway system, and the extreme line environment and varying driving speeds may affect drivers’ visual ergonomics. To elucidate the impact of the Snow and Ice (SI) line environment and different driving speeds on train drivers’ visual ergonomics, this paper introduces a novel research method. Utilizing a VR simulation driving scenario, the physiological parameters, task performance, and subjective evaluations of 44 participants were analyzed using IBM SPSS 26.0. The analysis covered conditions under two line environments (SI line and basic line) and three driving speeds (120 km/h, 160 km/h, and 200 km/h). The research reveals that the line environment, train speed, and their interaction effects significantly affect train drivers’ visual ergonomics in the following ways: (1) In the SI line environment, drivers experience significantly increased visual fatigue and discomfort, along with decreased visual performance. (2) With the increase of driving speed, drivers face heightened psychological pressure, shorter reaction times, and reduced accuracy. (3) The interaction effect between line environment and driving speed diminishes visual ergonomics by increasing physiological load and psychological stress, disrupting drivers’ emotional state and attention distribution, thus forming a negative cycle in train drivers’ visual ergonomics. The research findings provide extensive practical applications and customized recommendations for real-world driving scenarios.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"185 ","pages":"Article 106806"},"PeriodicalIF":4.7,"publicationDate":"2025-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143377223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Safety SciencePub Date : 2025-02-10DOI: 10.1016/j.ssci.2025.106807
Apurwa Dhoke, Pushpa Choudhary
{"title":"Pedestrian behavior under time pressure: A VR-based study of waiting duration and crashes at signalized intersection","authors":"Apurwa Dhoke, Pushpa Choudhary","doi":"10.1016/j.ssci.2025.106807","DOIUrl":"10.1016/j.ssci.2025.106807","url":null,"abstract":"<div><div>Pedestrians’ decision-making is significantly influenced by time pressure, potentially raising safety concerns. Therefore, this study seeks to quantify a comprehensive view of how pedestrians’ decision-making under time pressure affects waiting duration as a measure of traffic compliance. Further, it aims to examine the consequences of these decisions by estimating the probability of crashes under varying levels of time pressure. To understand pedestrian behavior while prioritizing safety, a virtual reality setup was utilized to record the movements of 60 participants at a four-legged signalized intersection. The experiments were conducted during three signal timing scenarios, including the end of the green, the middle of red, and the end of red, under three time pressure conditions: no time pressure, low time pressure, and high time pressure. To investigate the factors influencing waiting duration, a hazard-based duration model was used. Further, to identify the significant factors affecting the probability of crash occurrences, a generalized linear model was employed. The results show that waiting times significantly influence pedestrians’ decisions on whether to comply or violate signals, leading to increased urgency and unsafe crossing behavior under time pressure. Thus, time pressure reduces compliance behaviors, particularly during the end of the green signal phase. However, the crash probabilities were higher during the end of the red signal scenario, especially under high time pressure conditions. Ultimately, the current research findings identify an intricate relationship between pedestrians’ waiting behavior and the occurrence of crashes with the time pressure and provide some insightful recommendations for enhancing intersection safety.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"185 ","pages":"Article 106807"},"PeriodicalIF":4.7,"publicationDate":"2025-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143378582","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Safety SciencePub Date : 2025-02-09DOI: 10.1016/j.ssci.2025.106810
Jinghong Wang , Yuqing Liu , Qingzhao Li , Jialin Wu , Yan Wang , Juncheng Jiang
{"title":"Optimization of emergency shelter layout with consideration of toxic gas leakage based on a cell transmission model","authors":"Jinghong Wang , Yuqing Liu , Qingzhao Li , Jialin Wu , Yan Wang , Juncheng Jiang","doi":"10.1016/j.ssci.2025.106810","DOIUrl":"10.1016/j.ssci.2025.106810","url":null,"abstract":"<div><div>The layout and site selection of emergency shelters have a significant impact on the efficiency of large-scale crowd evacuation. However, during the actual evacuation process, there may be factors that affect the capacity of evacuation routes. This paper considers the problem of decreased road capacity due to toxic gas effects. By extending the Cell Transmission Model (CTM) and incorporating additional constraints, a linear programming model for emergency evacuation traffic was developed. Based on an improved CTM, this study examines the impact of different emergency shelter layouts in the Tianjin Port area. The results indicate that under the actual conditions at Tianjin Port, the evacuation efficiency differs by 22.43% between scenarios with and without the presence of toxic gases. A reasonable layout of additional shelters can improve evacuation efficiency by up to 45.33%. However, if the number of shelters is excessively increased, it could exacerbate congestion and reduce evacuation efficiency by up to 59.81% under the influence of toxic gases. The value of this study lies in its establishment of a model that can more accurately assess evacuation risks under the influence of toxic gases and inform the strategic placement of emergency shelters.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"185 ","pages":"Article 106810"},"PeriodicalIF":4.7,"publicationDate":"2025-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143372151","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Safety SciencePub Date : 2025-02-08DOI: 10.1016/j.ssci.2025.106809
Jianhua Zhang , Xiaohua Zhao , Ying Yao , Jushang Ou , Yujia Xiang
{"title":"The influence of individual characteristics and working environment on driving performance of truck drivers","authors":"Jianhua Zhang , Xiaohua Zhao , Ying Yao , Jushang Ou , Yujia Xiang","doi":"10.1016/j.ssci.2025.106809","DOIUrl":"10.1016/j.ssci.2025.106809","url":null,"abstract":"<div><div>Road freight transportation is a cornerstone of China’s economy, enabling efficient commodity circulation and network consumption. However, the high rate of truck-related accidents and fatalities is largely influenced by individual characteristics and company management practices. This study investigates the impact of these factors on driving safety using social cognitive theory as the framework. A questionnaire was designed to gather empirical data on truck drivers’ individual traits, work environments, and driving behaviors. Exploratory factor analysis confirmed the reliability and validity of the questionnaire. The association rule algorithm was then used to uncover correlations among drivers’ risk behaviors, personal attributes, and work environments. A structural equation model (SEM) was constructed, integrating findings from the association rule analysis and social cognitive theory to explore causal relationships and influence paths. This approach effectively addressed challenges in constructing optimal path models with multivariable relationships. The SEM results revealed that insomnia, negative affect, burnout, and safety climate significantly influence driving behavior. Mediation analysis further showed that well-being, weekly working days, and salary levels indirectly affect driving behavior through burnout and negative affect. These findings highlight critical targets for individual and organizational interventions. Freight companies can implement targeted education and training programs addressing risk factors and enhance safety management. This research offers practical insights for improving truck drivers’ behavior and promoting road freight safety.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"185 ","pages":"Article 106809"},"PeriodicalIF":4.7,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143372150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Safety SciencePub Date : 2025-02-08DOI: 10.1016/j.ssci.2025.106803
Mariella De Fino , Federica Cassano , Gabriele Bernardini , Enrico Quagliarini , Fabio Fatiguso
{"title":"On the user-based assessments of virtual reality for public safety training in urban open spaces depending on immersion levels","authors":"Mariella De Fino , Federica Cassano , Gabriele Bernardini , Enrico Quagliarini , Fabio Fatiguso","doi":"10.1016/j.ssci.2025.106803","DOIUrl":"10.1016/j.ssci.2025.106803","url":null,"abstract":"<div><div>The increasing use of VR (Virtual Reality) training tools in safety-related fields has prompted the scientific community to explore methods for assessing their effectiveness across different levels of immersion. While numerous studies have been conducted in sectors such as healthcare, transportation, agriculture, aviation, mining, firefighting, and construction, one area that remains underexplored is risk training for general public against natural and man-made disasters in both indoor and outdoor built environments. In this context, the paper aims at validating a prototype for VR multi-risk (heat wave and earthquake) training in urban open spaces that was developed according to a Serious Game (SG) approach. To address insights on its extensive adoption, the VR-SG was tested for comparison of three modes: traditional by video recording, non-immersive by desktop, immersive by headset. Outputs from feedback questionnaires on knowledge gain and transfer to expansive contexts, as well as on user experience, suggest that the effectiveness of knowledge acquisition differs significantly between headset and video, as well as between desktop and video, but shows no significant difference between headset and desktop. Additionally, while headset outperforms desktop in terms of engagement, perceived usefulness, and realism, it falls behind in ease of use. Finally, further analyses on training outcome by age, contents of open-ended answers, in-game errors and administration procedures supported the discussion of key aspects in targeting VR for public safety communication.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"185 ","pages":"Article 106803"},"PeriodicalIF":4.7,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143350310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Safety SciencePub Date : 2025-02-06DOI: 10.1016/j.ssci.2025.106805
Sharron O’Neill , Louise Thornthwaite
{"title":"Does legislating safety duties across the labour supply chain ensure contingent workers have equitable access to safety training?","authors":"Sharron O’Neill , Louise Thornthwaite","doi":"10.1016/j.ssci.2025.106805","DOIUrl":"10.1016/j.ssci.2025.106805","url":null,"abstract":"<div><div>Increasing demand for labour flexibility and the rise of the gig-economy has led increasingly to fractured labour supply chains. Recent legislative reforms in Australia and New Zealand place obligations on employers to ensure the health and safety of all workers whose work they control or direct, including, but not limited to employees. These reforms recognise the crucial role of OHS training. However, they expose an important gap in our understanding as to whether workers engaged in high-risk work across corporate boundaries now have equitable access to training.</div><div>This exploratory study surveyed 543 Australian permanent, causal and contract heavy vehicle drivers, comparing their access to various forms of OHS training and their experience and perceptions of hazardous events most commonly associated with fatal and high consequence injury in their industry. The reforms appear to improve access to external safety training courses, although significant differences in training participation mix and risk perceptions remain across employment types and some workers report multiple, often conflicting safety training messages.</div><div>The findings offer a novel and important contribution to the extensive literature on OHS training. Despite significant employer investment in training, efforts to absorb a cohort of (non-employee) workers into existing, corporate training systems appear largely unsuccessful and demonstrates a structural misalignment between (organisational) training models and (individuals’) training needs. However, resolving this mismatch will require a re-imagining of OHS training delivery to a more integrated model that can better identify and address gaps in the OHS knowledge and experience of an increasingly disparate group of workers.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"185 ","pages":"Article 106805"},"PeriodicalIF":4.7,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143208632","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Safety SciencePub Date : 2025-02-05DOI: 10.1016/j.ssci.2025.106802
Fatemeh Banani Ardecani, Amit Kumar, Sepehr Sabeti, Omidreza Shoghli
{"title":"Neural correlates of augmented reality safety warnings: EEG analysis of situational awareness and cognitive performance in roadway work zones","authors":"Fatemeh Banani Ardecani, Amit Kumar, Sepehr Sabeti, Omidreza Shoghli","doi":"10.1016/j.ssci.2025.106802","DOIUrl":"10.1016/j.ssci.2025.106802","url":null,"abstract":"<div><div>Work zone crashes and fatalities persist at alarming rates, highlighting the need for innovative safety solutions. Augmented Reality (AR) safety warnings have shown promise, yet their impact on situational awareness, attention, and cognitive workload remains underexplored. To bridge this gap, this study investigates the neurophysiological responses to AR-assisted warnings in roadway work zones under varying workload conditions. Leveraging electroencephalogram (EEG) technology within a virtual reality simulation of work zones, we objectively evaluated post-warning situational awareness, attention, and cognitive load during low-intensity (LA) and moderate-intensity (MA) activities. Key EEG indicators, including beta, gamma, alpha, and theta waves, as well as combined wave ratios, were used to measure these responses. Results revealed that AR warnings effectively triggered neurological responses associated with increased situational awareness and attention across both workload conditions. However, significant differences were observed in the timing and intensity of these responses. In the LA condition, peak responses occurred earlier (within 125 ms post-warning) and were more pronounced, suggesting a more robust cognitive response when physical demands were lower. Conversely, the MA condition showed delayed peak responses (125–250 ms post-warning) and more gradual changes, indicating a potential impact of increased physical activity on cognitive processing speed. These findings underscore the importance of considering physical workload when designing AR-based safety systems for roadway work zones. The research contributes to the understanding of how AR can enhance worker safety and provides insights for developing more effective, context-aware safety interventions in high-risk work environments.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"185 ","pages":"Article 106802"},"PeriodicalIF":4.7,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143156298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Safety SciencePub Date : 2025-02-05DOI: 10.1016/j.ssci.2025.106801
Grethe Midtlyng
{"title":"Just Enough for Resilience? Creating Capacity in a High-Security Prison","authors":"Grethe Midtlyng","doi":"10.1016/j.ssci.2025.106801","DOIUrl":"10.1016/j.ssci.2025.106801","url":null,"abstract":"<div><div>This article aims to investigate how human resources on the organisational front line can both challenge and support resilience in a prison system. Based on an ethnography of safety and security work in a Norwegian high-security prison, how operational senior officers manage resources on the spot to maintain safety and security is explored. The results show that senior officers created capacity as and for situated resilience through small acts of saving, stretching, and strengthening human resources. Due to their twofold position in both operations and management, senior officers seemed to hold a potential key function for counterbalancing organisational vulnerabilities through their embodied competencies. However, due to the limited available resources in the prison organisation, their management of resources often took the form of “firefighting”. Although sufficient in keeping specific parts of the prison safe and secure in specific situations, the achieved situated resilience did not necessarily contribute to avoiding or handling future disturbances. Moreover, because of the socially determined risk, the reallocation of resources could influence the existence and dimension of risk, due to reduced resources to prevent and mitigate beginning disturbances on the prison wings. These findings indicate that situated resilience, when not met with corresponding structural or systemic adjustments, can fuel organisational brittleness to future events. This research contributes to a bottom-up understanding of how situated resilience unfolds through the situated and embodied competency of operational managers, making the hidden but essential processes that constitute the order of a system more visible.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"185 ","pages":"Article 106801"},"PeriodicalIF":4.7,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143208634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Safety SciencePub Date : 2025-02-03DOI: 10.1016/j.ssci.2025.106804
Joana Eva Dodoo , Hosam Al-Samarraie , Ahmed Ibrahim Alzahrani , Tang Tang
{"title":"XR and Workers’ safety in High-Risk Industries: A comprehensive review","authors":"Joana Eva Dodoo , Hosam Al-Samarraie , Ahmed Ibrahim Alzahrani , Tang Tang","doi":"10.1016/j.ssci.2025.106804","DOIUrl":"10.1016/j.ssci.2025.106804","url":null,"abstract":"<div><div>The wider application of extended reality (XR) in various industrial settings has created numerous opportunities for enhancing worker safety. Several XR solutions have been applied to address specific safety challenges faced by workers. This study reviewed the current literature (2017–2024) on how XR technologies can potentially enhance worker safety. The PRISMA protocol was used to highlight how XR technologies are utilized in safety training for high-risk industries, their limitations, and recommendations for future improvements. Findings from a review of 41 studies indicate diverse opportunities (e.g., improved knowledge and productivity, delivery of interactive and sequential instructions) for virtual reality (VR), augmented reality (AR), and mixed reality (MR) in industries such as mining, construction, manufacturing, healthcare, power distribution/thermal plants, aviation, and firefighting. Several challenges (e.g., limited viewing fields, motion sickness, and control issues) were identified in the use of VR, AR, and MR, stemming from both human and socio-technical factors. The overall sentiment towards the use of XR in safety training was predominantly positive (550 instances), reflecting confidence in these technologies to enhance safety training outcomes. Findings from this study offer new insights into the capabilities of XR technologies in improving worker safety in high-risk industries and outline key considerations for policymakers and technology providers when integrating XR technologies to promote worker safety.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"185 ","pages":"Article 106804"},"PeriodicalIF":4.7,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143156320","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}