Risk Analysis最新文献

The objective of this study is to replicate the original study by Fischhoff et al. (1978) and its replication by Fox-Glassman and Weber (2016) and to examine whether risk perceptions for the previously studied activities and technologies have changed over the past 40 years, especially when activities/technologies related to contemporary concerns are included. To achieve this goal, the list of activities/technologies has been modified. To facilitate the analysis of individual data, all participants were asked to rate the benefits and risks of 24 activities. The within-participant approach was also used to achieve the second objective of our study: to analyze the relationship between perceived benefits and risks. In summary, the design of this study differed from previous studies in the following ways: (1) Nine activities/technologies were added related to contemporary concerns such as global warming and fake news on the Internet; (2) all participants rated both benefits and risks; (3) data were collected online (as in the 2016 study); (4) the study was conducted by Prolific with a sample size large enough to detect medium-size effects (n = 382). The two-factor structure proposed by Fischhoff et al.-unknown risk and dread risk-was confirmed on aggregated data for the new set of hazards, which included novel hazards. At the level of individual data, modest support for this structure was observed, and a very strong inverse relationship between perceived benefits and risks was observed.

The prediction of unmanned aerial vehicle (UAV) operators' unsafe acts is critical for preventing UAV incidents. However, there is a lack of research specifically focusing on UAV operators' unsafe acts, and existing approaches in related areas often lack precision and effectiveness. To address this, we propose a hybrid approach that combines the Human Factors Analysis and Classification System (HFACS) with random forest (RF) to predict and warn against UAV operators' unsafe acts. Initially, we introduce an improved HFACS framework to identify risk factors influencing the unsafe acts. Subsequently, we utilize the adaptive synthetic sampling algorithm (ADASYN) to rectify the imbalance in the dataset. The RF model is then used to construct a risk prediction and early warning model, as well as to identify critical risk factors associated with the unsafe acts. The results obtained through the improved HFACS framework reveal 33 risk factors, encompassing environmental influences, industry influences, unsafe supervision, and operators' states, contributing to the unsafe acts. The RF model demonstrates a significant improvement in prediction performance after applying ADASYN. The critical risk factors associated with the unsafe acts are identified as weak safety awareness, allowing unauthorized flight activities, lack of legal awareness, lack of supervision system, and obstacles. The findings of this study can assist policymakers in formulating effective measures to mitigate incidents resulting from UAV operators' unsafe acts.

The transition from one level of operations to a next larger, more complex level while maintaining coherence as a system has stymied organizational theorists for decades. Drawing on systems theory, network analysis, and collaborative governance, we explore how networks adapt during rapidly escalating crises. Specifically, we investigate the emergence of a synthesizing intelligence function among networks to support coordinated action. Using a case study of the 2020 Santa Clara Unit Lightning Complex Fire, we analyze field operations data from Incident Reports filed by the California Department of Forestry and Fire Protection to develop a system dynamics model. Our findings suggest that a synthesizing intelligence function, informed by various types of intelligence, influences the rate of change in operational systems during dynamic conditions. This system-wide intelligence function is crucial for decision-makers confronting extreme events, facilitating effective anticipation of complex transitions in large-scale operational systems.

In the residential sector, hurricane winds are a major contributor to storm-related losses, with substantial annual costs to the US economy. With the potential increase in hurricane intensity in changing climate conditions, hurricane impacts are expected to worsen. Current hurricane risk management practices are based on the hurricane risk assessment without considering climate impact, which would result in a higher level of risk for the built environment than expected. It is crucial to investigate the impact of climate change on hurricane risk to develop effective hurricane risk management strategies. However, investigation of future hurricane risk can be very time-consuming because of the high resolution of the models for climate-dependent hazard simulation and regional loss assessment. This study aims to investigate the climate change impact on hurricane wind risk on residential buildings across the southeastern US coastal states. To address the challenge of computational inefficiency, we develop surrogate models using machine learning techniques for evaluating wind and rain-ingress losses of simulated climate-dependent hurricane scenarios. We collect historical hurricane data and use selected climate variables to predict changing hurricane attributes under climate change. We build the surrogate loss model using data generated by the existing fragility-based loss model. The loss estimation of synthetic events using the surrogate model shows an accuracy with a 0.78 R-squared value compared to Hazard U.S. - Multi Hazard (HAZUS-MH) estimation. The results demonstrate the feasibility of utilizing surrogate models to predict risk changes and underline the increasing hurricane wind risk due to climate change.

This paper presents the design and implementation of an expert system for the domain of functional safety of machinery, featuring a novel multilingual chatbot interface developed using the Rasa framework. Unlike traditional expert systems, this approach aims to make the complex topic of functional safety more accessible to users with limited experience by assisting with tasks such as hazard identification, risk assessment, risk reduction, and safety function recommendation. The knowledge base of the system can be populated by functional safety experts through a graphical user interface, ensuring the system's utility and accuracy. This work demonstrates that the chatbot-based expert system retains many advantages of traditional expert systems while offering a more engaging user experience. An experimental evaluation of the presented expert system using hazard scenarios from real-life projects highlights the benefits of advanced machine learning techniques and pretrained embeddings, showing improvements in system performance. Continuous updates to the training dataset are essential for maintaining effectiveness in diverse environments. Compared to general-purpose chatbots like ChatGPT, this system provides reliable, standards-based insights. The system can be utilized by inexperienced machinery design personnel, such as mechanical and mechatronic engineers, before consulting with safety experts.

This study investigates how local context-specifically urban versus rural environments and socioeconomic conditions-influences individual crisis preparedness and resilience in Sweden. Using multilevel survey data from 12,574 respondents, we analyze both proactive preparedness actions and perceived resilience. Results show that rural residents report higher levels of preparedness and resilience than their urban counterparts. However, these differences in preparedness attenuate when controlling for individual risk perception, suggesting a mediating role. Socioeconomic context, on the other hand, does not show an independent effect beyond individual characteristics, indicating compositional rather than contextual influences. The findings highlight the importance of tailoring crisis preparedness strategies to both individual and local characteristics and stress the need for authorities to consider spatial disparities in vulnerability when planning for future crises.

The COVID-19 pandemic has exposed critical gaps in our management of systemic risks within complex, interconnected systems. This review examines 10 key areas where artificial intelligence (AI) and data analytics can significantly enhance pandemic preparedness, response, and recovery. Inadequate early warning systems, insufficient real-time data on resource needs, and the limitations of traditional epidemiological models in capturing complex disease dynamics are among the challenges analyzed. To address these issues, we explore the potential of AI applications, including machine learning-based surveillance, deep learning for improved epidemiological modeling, and AI-driven optimization of non-pharmaceutical interventions. These technologies offer the promise of more timely, accurate, and granular analysis of pandemic risks, thereby supporting evidence-based decision-making in rapidly evolving crises. However, implementing AI in pandemic response raises significant ethical and governance challenges, particularly concerning privacy, fairness, and accountability. We parse the promise and challenges of AI in the evolving space of emergency response data analytics and highlight critical steps forward.

Warning information plays a vital role in encouraging disaster preparedness among residents. Using survey experiment data from 619 respondents in China, this study examines how warning messages from AI, experts, and a combination of the two influence public disaster preparedness behaviors and whether the degree of impact differs between these sources. The findings reveal that warnings from AI, experts, and a combination of those two sources significantly affect disaster preparedness behaviors. Notably, and contrary to conventional expectations, the combined warnings from AI and experts do not result in a mutually strengthening effect. Instead, a crowding-out effect is observed, whereby the combined impact is less than the sum of individual effects ("Experts + AI < 2"). This outcome can be attributed to information fatigue, suggesting that information overload does not always benefit the public but instead often becomes a burden. Additionally, the influence of AI-driven warnings on preparedness varies substantially with respondents' educational levels. The insights provided by this study hold practical implications for government agencies in promoting public disaster preparedness.

This study employed the XGBoost model to conduct an in-depth analysis of consumer complaints and identified the key risk factors predicting vehicle recalls in the US market, providing valuable proactive risk management support for automakers and regulatory agencies. We leveraged the extensive data resources from National Highway Traffic Safety Administration to construct high-precision recall risk prediction models to predict the risk of recall. The models exhibited exceptional performance across different time windows, particularly maintaining a high level of area under the curve values over a prediction timespan of up to 18 months, demonstrating their predictive accuracy and stability. Our study contributes to risk management theory by addressing the challenges of integrating consumer complaints into predictive models for vehicle recall risk. While prior research has primarily focused on text mining of complaint content, our work systematically incorporates structured complaint data and recall records to enhance predictive accuracy. Also, our research distinguishes the indicators for the initial recall after launch to the market and the indicators for subsequent recalls, and bridges a critical gap in recall risk prediction at different stages of a vehicle's life cycle.

The vulnerability of critical networks to disinformation creates significant risks of disruption with potentially severe societal consequences. Maintaining secure and resilient networks, including infrastructure and supply chain networks, is important for ensuring economic productivity along with securing the health and well-being of society. An over-the-horizon threat to critical networks deals with adversaries who attack such networks indirectly by altering the consumption behavior of unwitting users influenced by weaponized disinformation. The proliferation of disinformation through various online platforms could pose a significant and evolving challenge able to compromise the resilience of critical networks. In this perspectives article, we review the literature in this area and offer some future research directions aimed at protecting networks from weaponized disinformation, enhancing their robustness, resilience, and adaptability.