Safety Science最新文献

{"title":"The three chairs: How the leaders of risk standards committees influence decision-making and accountability","authors":"Jan Hayes ,&nbsp;Martin Inge Standal ,&nbsp;Kristine Vedal Størkersen ,&nbsp;Sarah Maslen","doi":"10.1016/j.ssci.2025.107109","DOIUrl":"10.1016/j.ssci.2025.107109","url":null,"abstract":"<div><div>Standardization committees and their chairs sit at the center of standard development, and have influence over the standards’ scope, level of ambition, and framing. However, little is known about how the standard chairs themselves conceptualize and develop the standards. Situated in the literature on standardization and riskwork, we address what standard chairs see as the primary purpose of the standard for which they are responsible, how they conceptualize the usage of the standard, and the implications of their practice for risk management in general. We adopted a collated fieldwork approach, drawing together semi-structured interviews from two studies to look at the accounts of three standard chairs in Australia (AS 2885.6) and Norway (NORSOK Z-013 and NS 5814). We show how all chairs seek to influence the riskwork of direct users of the standards and senior managers. The chairs also emphasize the importance of expert judgement and reflecting on the inherent uncertainty in risk analyses. Chairs view the standards as frameworks rather than prescriptive methods. A good standard thus enables expert judgement, management decision making, and local adaptations while also establishing a rigorous process in risk management. This has implications for selection of standard chairs and for theoretical considerations of what it means to standardize.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"196 ","pages":"Article 107109"},"PeriodicalIF":5.4,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977377","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}

{"title":"Differential cognitive control in response to task severity during technologically-induced concurrent multitasking: The role of risk awareness in ICU nurses","authors":"Huiling Hu ,&nbsp;Tingting Feng ,&nbsp;Hui Ge ,&nbsp;Jiashuai Li ,&nbsp;Xue Wu ,&nbsp;Xuanna Wu","doi":"10.1016/j.ssci.2026.107112","DOIUrl":"10.1016/j.ssci.2026.107112","url":null,"abstract":"<div><h3>Objective</h3><div>To investigate how novice Intensive Care Unit (ICU) nurses engage cognitive control mechanisms when managing multitasking conflicts of differing severity, and how risk awareness modulates these processes.</div></div><div><h3>Background</h3><div>Digital transformation in healthcare has introduced more technological systems into ICUs, heightening multitasking demands and cognitive load. Novice nurses, lacking well-developed clinical schemas, rely more on executive control. Understanding how cognitive control is deployed—and how individual risk awareness shapes these processes—is essential for safeguarding patient safety in high-risk environments.</div></div><div><h3>Design</h3><div>EEG-based experimental study.</div></div><div><h3>Methods</h3><div>Twenty-six novice ICU nurses participated in a simulated technologically-induced multitasking paradigm. Behavioral accuracy, response times, and EEG measures (P1/P3 components, oscillatory power in alpha, beta, and theta bands) were recorded. Risk awareness was assessed via self-report scales.</div></div><div><h3>Results</h3><div>Nurses achieved higher accuracy in high-severity tasks, though response times did not differ significantly. ERP analyses showed significantly lower P1 peak amplitudes for high-severity tasks, particularly among nurses with higher risk awareness. Time–frequency analyses revealed greater alpha, beta, and theta power during low-severity tasks, suggesting increased cognitive flexibility under lower demands. Significant task severity × risk awareness interactions indicated that individuals with higher risk awareness deployed more targeted neural resources under high-severity conditions.</div></div><div><h3>Conclusions</h3><div>Task severity and individual risk awareness jointly shape cognitive control in technologically mediated multitasking among novice ICU nurses. These findings highlight the importance of integrating neurocognitive evidence into simulation-based training and interface design to strengthen nurses’ attentional management, reduce error risks, and enhance patient safety in the digital age of healthcare.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"196 ","pages":"Article 107112"},"PeriodicalIF":5.4,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977376","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}

{"title":"Describing psychosocial risk in the Mexican working population by sex, generation, and industry: a cross-sectional study","authors":"Courtney T Blondino ,&nbsp;Karla M Téllez ,&nbsp;Noémie Le Pertel ,&nbsp;Ariel Joab Almazan ,&nbsp;Lorna Friedman","doi":"10.1016/j.ssci.2025.107105","DOIUrl":"10.1016/j.ssci.2025.107105","url":null,"abstract":"<div><h3>Objectives</h3><div>Psychosocial risk or work-related hazard can lead to harmful individual and organizational outcomes. This study used 2023 data collected in compliance with Mexico’s NOM-035-STPS-2018 regulation to characterize overall psychosocial risk by sex, generation, industry, and explore domain-specific psychosocial risk by industry.</div></div><div><h3>Methods</h3><div>Frequencies (n) and percentages (%) were reported to describe the distribution of psychosocial risk. Significant difference testing was done with the chi-square statistic to test for differences in overall psychosocial risk by risk factor (α &lt; 0.05).</div></div><div><h3>Results</h3><div>Data from 58,994 employees from 67 organizations operating in Mexico that implemented NOM-035-STPS-2018 in 2023 were included. Approximately 27% of the sample had high or very high overall psychosocial risk. Males had significantly higher psychosocial risk than females (28.3% vs 25.5%, p &lt; 0.05) and there was no difference in psychosocial risk by generation. The construction, energy, and manufacturing industry had the highest level of psychosocial risk for lack of control over work (29.8%), leadership (20.5%), violence (12.8%), performance recognition (12.3%), working conditions (13.0%), and work relationships (3.5%) relative to the other four industry categories.</div></div><div><h3>Conclusions</h3><div>NOM-035 presents a unique opportunity to explore psychosocial risk in Mexico’s employees. In 2023, there was no difference in psychosocial risk by generation, and psychosocial risk was highest in workers of blue-collar industries. More research is needed to further explore these associations to inform interventions for employers and regulatory bodies.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"196 ","pages":"Article 107105"},"PeriodicalIF":5.4,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925864","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}

{"title":"Development of an evacuation transport model for residents in need of assistance in evacuation during nuclear disasters","authors":"Haruka Ohba ,&nbsp;Shinya Mizuno","doi":"10.1016/j.ssci.2026.107110","DOIUrl":"10.1016/j.ssci.2026.107110","url":null,"abstract":"<div><div>This study proposes a mathematical optimization model based on the Capacitated Vehicle Routing Problem to support the efficient transport of ”Residents in Need of Assistance in Evacuation” — such as the elderly and persons with disabilities — during nuclear disasters. The target area is Omaezaki City, Shizuoka Prefecture in Japan, where geographic open data — including census data, elevation data from the Geospatial Information Authority of Japan, and OpenStreetMap — were integrated to construct realistic road networks reflecting road accessibility during disasters. In particular, assuming tsunami-induced road flooding, we evaluated the impact of elevation-based road access restrictions on evacuation plans and identifying Residents in Need of Assistance in Evacuation in areas difficult to access. The model was optimized using both a Mixed-Integer Linear Programming approach and a genetic algorithm, and the results showed that Gurobi outperformed Genetic Algorithm in terms of both solution quality and computation time. The proposed model enables the quantitative identification of regions with limited accessibility based on the spatial distribution of home-based Residents in Need of Assistance in Evacuation and terrain characteristics, providing a practical decision support framework for local governments in disaster response planning and resource allocation.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"196 ","pages":"Article 107110"},"PeriodicalIF":5.4,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925863","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}

{"title":"The functional resonance analysis method in healthcare: How knowledge is produced within and across scientific communities","authors":"O. Lounsbury ,&nbsp;L. Pickup ,&nbsp;Riccardo Patriarca ,&nbsp;Marit S. de Vos ,&nbsp;Kate Preston ,&nbsp;Mark Sujan","doi":"10.1016/j.ssci.2025.107106","DOIUrl":"10.1016/j.ssci.2025.107106","url":null,"abstract":"<div><div>The Functional Resonance Analysis Method (FRAM) is a valuable tool for understanding and improving complex socio-technical healthcare systems. This Focused Mapping Review and Synthesis (FMRS) explores how the FRAM is used to produce knowledge and create insight about healthcare systems across three scientific communities: safety/engineering, human factors/cognitive science, and healthcare/health services research. We examined 33 included studies and identified key themes in how FRAM is applied within and across communities. The review highlights the critical role of knowledge brokers and boundary spanners in fostering interdisciplinary collaboration and knowledge transfer. There is limited epistemological clarity in most studies, which complicates cross-study comparisons and practical application. In addition, most studies are descriptive and do not develop actionable and robust interventions. We argue for greater transparency in epistemological positioning, methodological reflexivity, and the development of reporting guidelines to enhance the consistency and utility of FRAM studies. This review is a conceptual synthesis rather than a proof of concept. Future empirical studies should test whether explicit articulation of epistemological assumptions improves FRAM analyses, using the probes identified in this review as a starting point.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"196 ","pages":"Article 107106"},"PeriodicalIF":5.4,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925865","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}

{"title":"A maturity model for accident investigation: beyond technical and functional analysis","authors":"Riana Steen ,&nbsp;Stig O. Johnsen","doi":"10.1016/j.ssci.2025.107108","DOIUrl":"10.1016/j.ssci.2025.107108","url":null,"abstract":"<div><div>Causality is central to accident investigation, shaping how events are reconstructed, accountability is assigned, and systemic improvements are identified. However, conventional investigations often emphasise procedural and technical sequences, overlooking relational and interpretive dynamics crucial for understanding how systems drift towards failure through the gradual erosion of shared understanding. To examine what this framing may overlook, we revisit the 2018 collision between the Norwegian frigate Helge Ingstad and the tanker Sola TS. We first used FRAM to reconstruct Work-as-Imagined (WAI) and then conducted a sensemaking-informed content analysis to examine Work-as-Done (WAD) based on official reports. While FRAM effectively clarified system function coupling and where variability arose, it offered limited insight into how interpretations diverged or why shared understanding was difficult to sustain. This two-stage analysis revealed critical discrepancies between formal expectations and actual meaning-making. These insights, together with theoretical considerations, informed the development of a four-layer maturity model. The four-layer maturity model, consisting of (i) technical causality, (ii) human performance, (iii) socio-technical interdependencies, and (iv) relational sensemaking grounded in Human Readiness Levels (HRL). This model, with its distinctive fourth analytical layer (iv), shifts attention from functional interactions to how coherence forms—or fails—under pressure. The case was then revisited solely to illustrate how the model reveals relational and interpretive dynamics not captured by functional analysis, thereby avoiding methodological circularity. It highlights silence, saturation, and fragmentation as indicators of a system losing its capacity to adapt its understanding, even when information is available and routines continue.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"196 ","pages":"Article 107108"},"PeriodicalIF":5.4,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925862","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}

{"title":"From cumulative exposure to failure: a unifying modelling framework for nonlinear tissue fatigue in ergonomics","authors":"Jeff M. Barrett,&nbsp;Jack P. Callaghan","doi":"10.1016/j.ssci.2025.107099","DOIUrl":"10.1016/j.ssci.2025.107099","url":null,"abstract":"<div><div>Musculoskeletal injury risk is often assessed using models that assume damage accumulates linearly with loading. However, biological tissues may exhibit history-dependent changes in tolerance, particularly under repeated or variable loading. In this study, we propose a nonlinear cumulative damage model grounded in a mechanistic description of collagen fibril engagement and failure. The model predicts the evolving tissue tolerance over time, with damage rates governed by a Tobolsky-Eyring-type law modulated by prior damage history.</div><div>The model was calibrated using experimental fatigue data from functional spinal units and evaluated through a series of simulations designed to reflect common ergonomic exposures. These included constant-load cycling, variable-load sequences, and heavy-tailed loading distributions. Notably, the model predicts that tissue already compromised by prior loading is more susceptible to additional damage, even under identical external conditions—a form of path dependence not captured by the classical Miner-Palmgren rule.</div><div>Perturbation analysis reveals that commonly used fatigue models can be recovered as successive approximations of the proposed framework, offering a formal connection between linear cumulative load theory, including ergonomics tools like LiFFT, and our nonlinear formulation. This unifying perspective helps reconcile chronic and acute injury risk models and highlights the importance of accounting for load history and variability in injury risk assessments.</div><div>These findings suggest that ergonomic models should be sensitive not only to cumulative load, but also to its temporal structure and variability. Incorporating such nonlinearities could improve predictions of tissue failure and inform guidelines for safer task design.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"196 ","pages":"Article 107099"},"PeriodicalIF":5.4,"publicationDate":"2026-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884103","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}

{"title":"Mapping micro-mobility risk: AI-powered geospatial analysis and predictive modelling","authors":"Ömer Kaya ,&nbsp;Nuriye Kabakuş","doi":"10.1016/j.ssci.2025.107107","DOIUrl":"10.1016/j.ssci.2025.107107","url":null,"abstract":"<div><div>Micro-mobility vehicles have rapidly become widespread as a sustainable and practical alternative for urban transportation in recent years. In this study, micro-mobility vehicles refer to traditional bicycles, electric bicycles, and electric scooters, which represent the main categories of such modes involved in traffic crashes in Türkiye. Despite their growing popularity, the safety implications of these vehicles have not yet been fully understood, and comprehensive research addressing crash patterns and associated risk factors is required. To this end, this study employs an artificial intelligence-driven geospatial and statistical methodology. Crash reports involving micro-mobility vehicles in Türkiye between 2015 and 2023 were analysed. Seventeen independent variables and 102 sub-variables were identified and integrated into a GIS environment for spatial analysis. The impact levels of risk factors were assessed using six different Large Language Models (DeepSeek, GEMINI, Perplexity, ChatGPT, Copilot, and Poe). Crash risk maps and corresponding weight values were combined to produce an crash suitability map indicating the potential risk of micro-mobility crashes. Furthermore, the significance of these factors across different collision types was tested using a multinomial logistic regression model. To the best of the authors’ knowledge, this is the first study to apply a macro-scale dataset and an AI-enhanced geospatial decision-making approach to analyse micro-mobility crashes. The findings highlight the need for local governments and urban planners to implement targeted safety measures in regions with high crash potential.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"196 ","pages":"Article 107107"},"PeriodicalIF":5.4,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884104","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}

{"title":"The impact of data censorship on pre-movement time prediction in building fire evacuation: focusing on people with learning disabilities","authors":"Naser Kazemi Eilaki ,&nbsp;Trond Nordvik ,&nbsp;Carolyn Ahmer ,&nbsp;Ilona Heldal ,&nbsp;Håkan Frantzich ,&nbsp;Bjarne Christian Hagen","doi":"10.1016/j.ssci.2025.107103","DOIUrl":"10.1016/j.ssci.2025.107103","url":null,"abstract":"<div><div>In fire evacuations, pre-movement time (PT) is an important part of the total egress time. Data from fire drills is commonly used to investigate occupant behavior, including PT, but these data are often subjected to censoring. This study examines the impact of data censoring on PT analysis during fire evacuation drills, which is important for predicting overall evacuation time. It identifies various types of data censorship and highlights the importance of accounting for censored data in statistical analyses. The research utilizes the Accelerated Failure Time model (AFT), a parametric survival analysis method, to address data censoring issues. Data from two fire evacuation drills in residential buildings with occupants with learning disability were analyzed using two scenarios: one excluding censored data and another incorporating it. The results demonstrate that accounting for censored data significantly alters the interpretation of the causal model, with covariates such as fire incident time emerging as significant only when censored data is considered. The AFT model effectively manages censored data by updating predictions for observed events and extrapolating values for censored observations using the best-fit distribution. This study underscores the importance of incorporating censored data into evacuation models emphasizing that excessive data censoring reduces the validity of AFT model predictions. Although limited by sample size, these findings offer valuable insights into effective covariates influencing PT and provide guidance for future evacuation modeling and simulation tools.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"196 ","pages":"Article 107103"},"PeriodicalIF":5.4,"publicationDate":"2025-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884102","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}

{"title":"Artificial intelligence in hospital fall Prevention: Current Applications, Challenges, and Future Directions","authors":"Adewoyin A. Osonuga ,&nbsp;Ayokunle Osonuga ,&nbsp;Deborah Omeni ,&nbsp;Gloria C. Okoye ,&nbsp;Eghosasere Egbon ,&nbsp;David B. Olawade","doi":"10.1016/j.ssci.2025.107104","DOIUrl":"10.1016/j.ssci.2025.107104","url":null,"abstract":"<div><div>Hospital falls represent a critical patient safety challenge, affecting millions of patients globally and resulting in substantial morbidity, mortality, and healthcare costs. Traditional fall prevention strategies, whilst beneficial, often lack the precision and real-time responsiveness needed for optimal patient protection. This narrative review with systematic search examines the current applications, effectiveness, and implementation challenges of artificial intelligence (AI) technologies in hospital fall prevention. A comprehensive search was conducted across PubMed, EMBASE, IEEE Xplore, and Google Scholar databases from January 2015 to October 2024. AI technologies demonstrate promise across four primary domains: machine learning predictive models achieving AUROC of 0.85–0.97 (with calibration reported variably), computer vision systems enabling real-time behavioural monitoring (94–97% detection accuracy in controlled settings), sensor-based technologies providing continuous patient surveillance (89–96% accuracy with multi-sensor fusion), and natural language processing enhancing risk factor extraction from clinical documentation (sensitivity 95% CI in selected studies). These metrics represent primarily single-site, retrospective studies with limited external validation and variable baseline fall rates. Successful implementations report fall reduction rates of 0.9–1.2 falls per 1,000 patient-days (15–40% relative reduction) across various healthcare settings, though baseline rates ranged from 2.8 to 5.1 falls per 1,000 patient-days across different care settings, and secular trends and study design heterogeneity limit causal inference. AI-driven systems offer enhanced prediction accuracy, real-time monitoring capabilities, and personalised risk assessment. However, implementation challenges include alarm fatigue (alert rates and positive predictive value rarely reported), algorithmic bias requiring ongoing fairness audits, liability concerns when AI systems fail to prevent falls, data privacy concerns, integration complexities, clinical workflow adaptation, and substantial cost barriers for smaller institutions. Future developments should prioritize explainable AI systems, multisite external validation with standardised metrics (AUROC, AUPRC, calibration), federated learning approaches, and implementation trials examining both fall rates and care process outcomes.</div></div>","PeriodicalId":21375,"journal":{"name":"Safety Science","volume":"196 ","pages":"Article 107104"},"PeriodicalIF":5.4,"publicationDate":"2025-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145840671","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}