Dynamic Risk Research Articles

Characteristics of Overburden Damage and Rainfall-Induced Disaster Mechanisms in Shallowly Buried Coal Seam Mining: A Case Study in a Gully Region

Shallow coal mining in gully regions has resulted in significant subsidence hazards and increased the risk of surface water inflow into mining panels, compromising the sustainability of surface water management and underground resource exploitation. In this study, the chain disaster process caused by shallow coal seam mining and heavy rainfall is quantitatively analyzed. The findings reveal that shallow coal seam mining leads to the formation of caved and fractured zones in the vertical direction of the overlying rock. The fractured zone can be further classified into a compression subsidence zone and a shear subsidence zone in the horizontal direction. The shear subsidence zone is responsible for generating compression and shear deformations, intercepting rainfall runoff, and potentially triggering landslides, necessitating crack landfill treatments, which are critical for promoting sustainable mining practices. The HEC-RAS program was utilized to integrate annual maximum daily rainfall data across different frequencies, enabling the establishment of a dynamic risk assessment model for barrier lakes. Numerical simulations based on unsaturated seepage theory provide insights into the infiltration and seepage behavior of rainfall in the study area, indicating a significant increase in saturation within lower gully terrain. Precipitation infiltration was found to enhance the saturation of the shallow rock mass, reducing matric suction in unsaturated areas. Finally, the disaster chain is discussed, and recommendations for managing different stages of risk are proposed. This study offers a valuable reference for the prevention and control of surface water damage under coal mining conditions in gully regions.

Dynamic risk assessment of Uncertain Random System considering operator's simple emergency-stop action in short time window

Human-machine systems can be considered as typical Uncertain Random Systems (URS). In human-machine systems, operator's simple emergency-stop actions act as a vital soft barrier. Conventional Human Reliability Analyses face challenges in evaluating operator's simple emergency-stop action due to their static characteristics, incompatibility with short time windows, and problems posed by small samples. This paper provides a novel solution to risk assessment of human-machine systems involving operator's simple emergency-stop mechanism. Specifically, an evaluation method of operator's simple emergency-stop action is established, where human cognition is decomposed according to its temporal and logical dimensions, key indicators of human sub-cognition are evaluated, and uncertainty theory is used to integrate these indicators. The output of proposed method is uncertainty of operator's behavior (UOB), which can offer more information than traditional Human Error Probability (HEP) and can be readily converted into uncertainty of human error. Further, a precise simulation method for dynamic risk assessment of Uncertain Random Systems is developed, in which the integration of the UOB is achieved. A case study demonstrates the effectiveness of the proposed methods, and its results show that the proposed methods can clearly reflect the differences in system risk under short time window conditions that have minor distinctions.

A hierarchical multi-parametric programming approach for dynamic risk-based model predictive quality control

In this work, we present a hierarchical batch quality control strategy with real-time process safety management. It features a multi-time-scale decision-making framework augmenting: (i) Risk-aware model predictive controller for short-term set point tracking and dynamic risk control under disturbances; (ii) Control-aware optimizer for long-term quality and safety optimization over the entire batch operation; (iii) Intermediate surrogate model to bridge the timescale gap by readjusting the optimizer operating decisions for the controller. All of the above problems are solved via multi-parametric mixed-integer quadratic programming with a key advantage to generate offline explicit control/optimization laws as affine functions of process and risk variables. This allows for the design of a fit-for-purpose risk management plan prior to real-time implementation, while reducing the need for repetitive online dynamic optimization. A unified process model is used to underpin the consistency of hierarchical operational optimization. The proposed approach offers a flexible strategy to integrate distinct decision-making time scales which can be selected separately tailored to the process-specific need of control, fault prognosis, and end-batch quality control. A T2 batch reactor case study is presented to showcase this approach to systematically address the interactions and trade-offs of multiple decision layers toward improving process efficiency and safety.

Dynamic risk analysis of tank cleaning operations using bow-tie-based fuzzy Bayesian network

This study aims to enhance maritime safety by focusing on risk analysis in tank cleaning operations, specifically targeting the risk of spillages. The Bow-Tie-based Fuzzy Bayesian Network (BT-FBN) methodology was employed to conduct an equipment-based dynamic risk analysis. A real chemical tanker was selected as the model vessel to ensure realistic calculations, and ‘Loss of Containment’ (LoC) was identified as the critical event for comprehensive risk analysis. The BT methodology was utilised to identify and systematically map potential hazards. Considering the temporal dynamics of risk, a Bayesian dynamic risk analysis was performed using four-month tank cleaning data from the model vessel, and fuzzy logic was applied to address uncertainties in the data. The findings indicated a steady increase in the risk of ‘LoC’ from 4.33% to 11.86% over four months, consistent with observed accident rates (8.57%). Furthermore, the likelihood of consequential events, such as ‘Delay in Operations & Near Miss’ (C-1), showed a similar trend, ranging from 3.69% to 10.09%. Notably, the analysis underscored the critical role of ‘Human Error & Management Failure’ (IE4) in contributing most significantly to ‘LoC’, ranging from 2.86% to 8.77%, emphasising the importance of addressing human factors to mitigate environmental pollution.

Dynamic Risk Forecasting Based on Deep Learning and Collapse Risk Comprehensive Evaluation of Mountain Tunnel Portal Construction

Dynamic Risk Forecasting Based on Deep Learning and Collapse Risk Comprehensive Evaluation of Mountain Tunnel Portal Construction

On [formula omitted]-expectations and filtration-consistent nonlinear expectations

In this paper, we obtain a comparison theorem and an invariant representation theorem for backward stochastic differential equations (BSDEs) without any assumption on the second variable z. Using the two results, we further develop the theory of g-expectations. Filtration-consistent nonlinear expectation (F-expectation) provides an ideal characterization for the dynamical risk measures, asset pricing and utilities. We propose two new conditions: an absolutely continuous condition and a (locally Lipschitz) domination condition. Under the two conditions respectively, we prove that any F-expectation can be represented as a g-expectation. Our results contain a representation theorem for n-dimensional F-expectations in the Lipschitz case, and two representation theorems for 1-dimensional F-expectations in the locally Lipschitz case, which contain quadratic F-expectations.

A fuzzy logic-based risk assessment framework for the crude oil transportation supply chain

This study proposes a novel approach to increase the efficiency of crude oil transportation by integrating fuzzy logic and the TOPSIS method. The proposed technique addresses the uncertainties and imprecisions in decision-making processes to identify the optimal shipping alternatives. The study also evaluates the static and dynamic risk indicators related to the transportation of crude oil. The China-Pakistan Economic Corridor (CPEC) is found to be a promising option for crude oil shipments, reducing the risks associated with current routes. Using a multi-criteria decision-making (MCDM) methodology, the research determines China's most efficient crude oil supply channel using a fuzzy-based Improved TOPSIS approach. The economic feasibility of the CPEC option has been confirmed through a thorough cost-benefit analysis (CBA). Additionally, the optimal mode of transportation from CPEC seaports to Western China is determined using the fuzzy-based Improved TOPSIS method. The MCDM criteria considered in all scenarios include time, financial outlays, energy consumption, environmental impact, and security risks. The results indicate that the CPEC route and the maritime route via Myanmar outperform the conventional route that passes through the seaports of Eastern China. This study emphasizes the significant advancement in decision-making in crude oil transportation made possible by the fuzzy-based Improved TOPSIS technique.

Family Dynamics, Socioeconomic Hardships, and Health Risk Behaviours of Bulgarian Adolescents during the COVID-19 Pandemic.

This study aims to explore family dynamics and the economic hardships experienced by families during the COVID-19 pandemic and their associations with adolescents' health risk behaviours (HRBs). Based on a representative study of adolescents aged 11-16 conducted in Bulgaria during the COVID-19 pandemic and HBSC data from the pre-pandemic period, logistic regression models were applied, assessing cigarette smoking, vaping, alcohol use, drunkenness, and cannabis use. The independent variables included demographics, Family Affluence Scale (FAS III), family structure, ease of communication with parents, and the authors' developed questions on parents' income and economic status change, family conflicts, and missing contact with extended family due to the pandemic. Material status of the family showed increasing differentials in adolescents' HRBs during the pandemic. Parental unemployment, income reduction, and temporary lay-offs were associated with a higher risk of substance use. Family conflicts, missing contact with extended family, and difficulties in communication with the mother were related to a higher risk of substance use. Communication with the father was significantly associated with alcohol use and drunkenness. Boys had lower odds of vaping and higher odds of alcohol use, drunkenness, and cannabis use. Higher age and minority status were associated with an increase in adolescents' HRBs. This study highlights the need for special family-focused interventions in times of health and economic crises.

Pediatric Papillary Thyroid Carcinoma: Outcomes After Surgery Without Adjuvant Radioactive Iodine.

Pediatric papillary thyroid carcinoma (PTC) is usually treated with total thyroidectomy followed by radioactive iodine (RAI). Recently, RAI is being used more selectively based on surgical pathology and postoperative dynamic risk stratification (DRS). To describe patients with pediatric PTC not initially treated with RAI and their disease outcomes. This was an ambispective study at a tertiary cancer center of patients < 19 years diagnosed from 1/1/1990 to 12/31/2021 with stage I PTC who intentionally were not treated with RAI within a year of diagnosis. We assessed clinical characteristics, management, and disease outcomes using DRS. Of 490 PTC patients, we identified 93 eligible patients (median age at diagnosis 16y; 87% female), including 46 (49%) with cervical lymph node metastases. Initial management included: total thyroidectomy ± neck dissection (n=69, 75%), lobectomy ± neck dissection (n=20, 21%), or a Sistrunk procedure for ectopic PTC (n=4, 4%). After a median follow-up of 5.5 years (range 1-26), most patients (85/93; 91%) remained disease-free with no further therapy. Persistent (n=5) or recurrent (n=3) disease was found in 9% of the entire cohort. Four patients ultimately received RAI, of which only one clearly benefited, and additional surgery was performed or planned in four patients, two of whom had an excellent response at last follow-up. Selected pediatric PTC patients, even those with lymph node metastases, may not require therapeutic 131I and can avoid the unnecessary risks of RAI while still benefitting from the excellent long-term outcomes that are well-described for this disease.

Dynamic risk assessment of natural gas transmission pipelines with LSTM networks and historical failure data

In the realm of energy infrastructure, ensuring the security of gas transmission pipelines is critical. This research introduces an advanced dynamic risk assessment framework that leverages the predictive capabilities of LSTM networks, presenting an improvement over conventional failure prediction models. Unlike traditional approaches that rely on averaging historical failure records, this framework dynamically processes historical pipeline failure incidents into sequential time series analysis, facilitating and improving the accuracy of the current failure rate. The model refines the failure rate estimation for individual pipelines by incorporating unique characteristics and modification factors, resulting in a highly precise failure likelihood estimation. Additionally, this study introduces a quantifiable linkage between mortality risk and the fatality probit value across various accident scenarios, enhancing consequence evaluation. A sensitivity analysis is then performed to assess the impact of various input parameters on the model's performance. The practical application of the model on a U.S. pipeline confirms its effectiveness. This proposed methodology substantially enhances the understanding of incident causation in gas pipeline systems, paving the way for superior safety management strategies. It is instrumental in enhancing pipeline safety, refining infrastructure planning, and optimizing safety resource allocation. The methodology offers benefits for pipeline operators, industry professionals, and regulatory agencies, contributing to improved operational safety and resource management in the pipeline industry.

Mercury Dynamics and Bioaccumulation Risk Assessment in Three Gold Mining-Impacted Amazon River Basins.

Mercury contamination from gold mining in the Amazon poses significant environmental and health threats to the biome and its local populations. The recent expansion of non-industrial mining areas has severely impacted territories occupied by traditional communities. To address the lack of sampling data in the region and better understand mercury dynamics, this study used the probabilistic model SERAFM to estimate the mercury distribution and bioaccumulation in fish. The analysis covered 8,259 sub-basins across three major Amazonian basins: the Branco, Tapajós and Xingu rivers. The findings revealed increasing downstream mercury levels, with notable accumulations in the main watercourses influenced by methylation processes and mining releases. The projected concentrations showed that an average of 27.47% of the sub-basins might not comply with Brazilian regulations, rising to 52.38% in the Branco and Tapajós river basins separately. The risk assessment of fish consumption based on the projections highlighted high mercury exposure levels among traditional communities, particularly indigenous populations, with an average of 49.79% facing an extremely high risk in the Branco and Tapajós river basins. This study demonstrated SERAFM's capacity to fill information gaps in the Amazon while underscoring the need for enhanced data collection, culturally sensitive interventions and regulatory updates to mitigate mercury contamination in gold mining-affected areas.

Dynamic connectedness of climate risks, oil shocks, and China’s energy futures market: Time-frequency evidence from Quantile-on-Quantile regression

This study investigates the dynamic risk nexus among climate risks, oil shocks and China’s energy futures market from a time–frequency-quantile perspective. We first explore the dynamic connectedness of “climate risks – oil shocks – energy futures” and examine the risk transmission channels through mediation effects model. The Quantile-on-Quantile regression is used to study the time–frequency impact of climate risks and oil shocks on energy futures across different market conditions and investment horizons. Our empirical results are as follows: First, climate transition risks, oil demand and risk shocks play mediating roles in risk transmission channels. Second, the impact of climate risks and oil shocks on energy futures is heterogeneous and asymmetric under extreme conditions. Notably, global warming, oil supply shock and international climate summits are the greatest shocks to China’s energy market. Finally, climate risks and oil shocks are more pronounced in the short term. Overall, these findings offer valuable insights for shaping risk management strategies and implementing effective hedging practices within the energy market.

Data-driven credit risk monitoring: Leveraging machine learning in risk management

This review explores the evolution of credit risk monitoring, tracing its journey from traditional qualitative assessments to the current integration of machine learning (ML). It highlights how the integration of ML and big data has introduced unprecedented capabilities for analyzing extensive datasets and detecting subtle patterns beyond human capacity. These advanced technologies enable more accurate, efficient and dynamic credit risk predictions through techniques such as random forests, gradient boosting and decision trees. The transformative potential of these methodologies in credit risk assessment was critically examined, addressing challenges such as legacy system integration, data quality and regulatory compliance. It emphasizes the importance of incorporating forward-looking macroeconomic indicators to comply with applicable financial reporting standards and regulatory requirements. Furthermore, it highlights the necessity of ensuring model transparency to maintain trust and compliance. By leveraging the power of big data and ML, the study shows that financial institutions can achieve more precise and proactive risk assessments, enhancing decision-making processes and mitigating potential risks. This comprehensive review provides valuable insights for stakeholders, guiding the implementation of advanced analytical techniques to improve credit risk management. Ultimately, it underscores the potential for a more robust, efficient and stable financial system through the strategic application of ML and big data analytics. Keywords: Credit Risk Management, Financial Services, Machine Learning, Big Data, Financial Transformation.

Longitudinal tracing of on-treatment plasma Epstein-Barr virus DNA as a biomarker for real-time dynamic risk monitoring in patients with nasopharyngeal carcinoma: The EP-SEASON study.

136 Background: Plasma Epstein-Barr virus (EBV) DNA is a biomarker for nasopharyngeal carcinoma (NPC). However, whether longitudinal on-treatment EBV DNA tracing would inform real-time recurrence risk in NPC remains to be elucidated. To address this issue, we conducted EP-SEASON (NCT03855020), a prospective cohort study of longitudinal on-treatment EBV DNA in EBV-positive NPC patients. Methods: Non-metastatic NPC with detectable pretreatment EBV DNA that received standard sequential chemo-radiotherapy were recruited. Serial EBV DNA in blood specimens were collected across pretreatment, 3-week after each induction chemotherapy (IC), every-week during radiotherapy (RT), within 1-week and 1-3 months upon RT completion. EBV DNA extracted from plasma was quantified by a real-time quantitative polymerase chain reaction assay targeting the BamHI-W region of the EBV genome. Longitudinal changing patterns of EBV DNA and their clinical relevance were analyzed. Results: Between May 2019 and April 2021, a total of 1000 patients underwent per-protocol EBV DNA assessments and treatment; 248 and 752 patients received concurrent chemoradiotherapy (CCRT, subcochortno-IC) and IC plus CCRT (subcochortIC), respectively. Longitudinal changes and clearance rates of EBV DNA displayed unique patterns during IC and RT phases. Throughout IC and RT phases, the rate of EBV DNA clearance decreased with repeated IC cycles; interestingly, the rate re-rose upon the introduction of RT, and then dropped to low levels (&lt;10%/week) after the 4th week of RT (post-RT4w). Patients in subcochortno-IC and subcochortIC also exhibited differential clearance patterns during RT, with the rate of clearance being higher in subcochortno-IC before the 2nd week of RT (post-RT2w). Notably, despite the prognostic significance of EBV DNA at each timepoint ( Pall &lt; 0.05), longitudinal EBV DNA displayed a linkage changing map with dynamic recurrence risks: 3-year disease-free survival (DFS) rate steadily dropped with delayed EBV DNA clearance during IC; it re-rose at the 2nd RT week upon EBV DNA clearance, followed by a sharp decrease at the 4th RT week in subcochortIC. In contrast, 3-year DFS rate continuously decreased with delayed EBV DNA clearance in subcochortno-IC. Finally, we identified different phenotypes of whole-course circulating tumor DNA changing dynamics that showed distinct prognosis ( P &lt; 0.05). Conclusions: Longitudinal on-treatment EBV DNA tracing could forecast real-time risks and facilitate risk-adapted decision-making in NPC.

Application of machine learning to predict in-hospital mortality after transcatheter mitral valve repair

IntroductionTranscatheter mitral valve repair offers a minimally invasive treatment option for patients at high risk for traditional open repair. We sought to develop dynamic machine-learning risk prediction models for in-hospital mortality after transcatheter mitral valve repair using a national cohort. MethodsAll adult hospitalization records involving transcatheter mitral valve repair were identified in the 2016–2020 Nationwide Readmissions Database. As a result of initial class imbalance, undersampling of the majority class and subsequent oversampling of the minority class using Synthetic Minority Oversampling TEchnique were employed in each cross-validation training fold. Machine-learning models were trained to predict patient mortality after transcatheter mitral valve repair and compared with traditional logistic regression. Shapley additive explanations plots were also developed to understand the relative impact of each feature used for training. ResultsAmong 2,450 patients included for analysis, the in-hospital mortality rate was 1.8%. Naïve Bayes and random forest models were the best at predicting transcatheter mitral valve repair postoperative mortality, with an area under the receiver operating characteristic curve of 0.83 ± 0.05 and 0.82 ± 0.04, respectively. Both models demonstrated superior ability to predict mortality relative to logistic regression (P < .001 for both). Medicare insurance coverage, comorbid liver disease, congestive heart failure, renal failure, and previous coronary artery bypass grafting were associated with greater predicted likelihood of in-hospital mortality, whereas elective surgery and private insurance coverage were linked with lower odds of mortality. ConclusionMachine-learning models significantly outperformed traditional regression methods in predicting in-hospital mortality after transcatheter mitral valve repair. Furthermore, we identified key patient factors and comorbidities linked with greater postoperative mortality. Future work and clinical validation are warranted to continue improving risk assessment in transcatheter mitral valve repair .

JmBIG: enhancing dynamic risk prediction and personalized medicine through joint modeling of longitudinal and survival data in big routinely collected data

We have introduced the R package jmBIG to facilitate the analysis of large healthcare datasets and the development of predictive models. This package provides a comprehensive set of tools and functions specifically designed for the joint modelling of longitudinal and survival data in the context of big data analytics. The jmBIG package offers efficient and scalable implementations of joint modelling algorithms, allowing for integrating large-scale healthcare datasets.By utilizing the capabilities of jmBIG, researchers and analysts can effectively handle the challenges associated with big healthcare data, such as high dimensionality and complex relationships between multiple outcomes.With the support of jmBIG, analysts can seamlessly fit Bayesian joint models, generate predictions, and evaluate the performance of the models. The package incorporates cutting-edge methodologies and harnesses the computational capabilities of parallel computing to accelerate the analysis of large-scale healthcare datasets significantly. In summary, jmBIG empowers researchers to gain deeper insights into disease progression and treatment response, fostering evidence-based decision-making and paving the way for personalized healthcare interventions that can positively impact patient outcomes on a larger scale.

Dynamic risk evolution analysis for in-situ combustion incidents of offshore heavy oil recovery

In-situ combustion for offshore heavy oil involves significant risk factors related to ignition, gas injection, and corrosion prevention, especially under challenging operating conditions. If not properly managed, these risks can lead to explosions in injection or production wells. This paper proposes a hybrid HDBN-based risk analysis method to estimate the probability of explosions during the three critical phases of in-situ combustion and to diagnose key risk factors contributing to such incidents. The method introduces new bootstrap terms to in-situ combustion and constructs a hybrid DBN to capture the dynamic behavior of explosion incidents. It incorporates the Delphi method and triangular Pythagorean fuzzy theory to address data scarcity and uncertainty through affiliation and uncertainty functions. A key aspect of the proposed method is the graded quantification of damage to the casing due to corrosion and thermal stresses. A case study of a well in Liaodong Bay, Bohai Sea, demonstrates the feasibility of this method. The study captures dynamic risk evolution characteristics during three stages of in-situ combustion and conducts sensitivity analysis to obtain key risk factors such as "bottom high temperature" and "gas corrosion".

Dynamic credit risk transmissions among global major industries: Evidence from the TVP-VAR spillover approach

We examine the dynamics of credit risk connectedness by analyzing credit default spreads in four major sectors (banks, transportation, manufacturing, electricity) across three global regions (Asia, Europe, North America) using the TVP-VAR spillover methodology from 2007 to 2024. We have identified significant findings regarding credit risk spillovers among them. First, there are consistently high levels of credit risk spillovers between sectors, indicating underlying economic factors influencing the transmission of credit risk shocks. Second, notable regional findings include a substantial increase in credit risk connectedness for Asian banks during the global financial crisis (GFC). European manufacturing sectors also displayed significantly high connectedness levels during both the GFC and the COVID-19, while North American banks saw a notable surge due to the collapse of Silicon Valley Banks (SVB) in March 2023. In addition, during the RU-war, the electricity and manufacturing sectors in Europe had high CDS connectedness. Lastly, a distinct observation emerged concerning the Asian transportation sector. These findings have practical implications for policymakers and portfolio managers. For instance, they can help policymakers assess the effectiveness of their policies by revealing global industry credit risk interconnections. Additionally, the dynamic credit risk linkages provide strategies for hedging credit risk.

Assessment of dynamic hydrological drought risk from a non‐stationary perspective

AbstractThe stationarity hypothesis of hydrometeorological elements has been questioned in the context of global warming and intense human disturbance. The conventional drought index and methods of frequency analysis may no longer be applicable for hydrological drought risk evaluation under a changing environment. In this study, a new dynamic hydrological drought risk evaluation framework is proposed for application to the Hanjiang River basin (HRB), which simultaneously considers the non‐stationarity in the construction of drought index as well as in the frequency analysis. First, a non‐stationary standardized runoff index (NSRI) is developed using a generalized additive model for location, scale and shape (GAMLSS) framework. Then, hydrological drought characteristics including duration and severity are identified, and their marginal distributions are established. Finally, based on the dynamic copula, considering the non‐stationarity of the dependence structure, the dynamic joint probability distribution, conditional probability distribution and return period of the bivariate hydrological drought properties are analysed. Results showed that NSRI, which integrates the impacts of climate change and anthropogenic activities on the non‐stationarity of runoff series, had a better ability to capture runoff extremes than had SRI. In addition, it is indispensable to consider the non‐stationarity of the dependence structure between variables when discussing the multivariate joint risk of hydrological drought. The risk of hydrological drought in the study area has shown an increasing trend in the past 65 years, and the drought conditions from upstream to downstream have been alleviated first and then intensified. This study provides valuable information for regional drought risk estimation and water resources management from a non‐stationary perspective.

Landslides - Multi-Hazards Dynamics and Disaster Risk Reduction Through Integrated Participatory Community Approaches

Landslides - Multi-Hazards Dynamics and Disaster Risk Reduction Through Integrated Participatory Community Approaches