Bayesian Risk Model Research Articles

Smart meter privacy control strategy based on multi‐agent hidden Markov energy management model under low trust communication

AbstractWith the popularity of smart meters, the frequent information exchange between smart grids and consumers leads to easy leakage of consumers' electricity consumption data. These leaked electricity consumption data are obtained by some malicious attackers and used to infer consumers' behavioural patterns by non‐intrusive load monitoring (NILM), which seriously threatens consumers' privacy. Therefore, the multi‐agent Hidden Markov energy management model is proposed in this paper to safeguard the privacy of consumers. First, a weighted Bayesian risk model is proposed, which combines privacy leakage risks and energy storage system (ESS) losses in a microgrid with multiple agents. Next, a three‐loop model for lithium batteries is constructed to quantify the capacity degradation and cost issues of the ESS. Finally, the multi‐objective optimization problem is resolved by integrating the Bayesian risk model with a hidden Markov model to simulate attackers. The proposed multi‐agent Markov decision process method is validated on Electricity Consumption and Occupancy (ECO) dataset, and control strategies are evaluated based on different weights in the Bayesian risk model. The results demonstrate that by incorporating the multi‐agent approach and energy storage system capacity degradation into the privacy protection strategy, the lifespan of the energy storage system can be significantly increased.

An Integrated Bayesian Risk Model for Coastal Flow Slides Using 3-D Hydrodynamic Transport and Monte Carlo Simulation

The literature suggests two forms of flow slides: breaching and liquefaction. Both forms of failure have comparable ultimate circumstances, but the progression and sand movement mechanisms of breaching failure diverge from those of liquefaction. The first type, breaching, occurs in densely packed sand and is characterized by slow sand grain discharge throughout the dilation of the failing soil particles and negative excess pore pressures. The latter form, known as liquefaction, is the process by which a mass of soil abruptly begins to behave like a flowing liquid, and as a result, it can flow out across overly mild slopes. The process begins in compacted sand and is linked to positive surplus pore water pressures that are caused by the compaction of the sand. Despite the available literature on flow slide failures, our understanding of the mechanisms involved remains limited. Since flow slides often begin below the water surface, they can go undetected until the collapse reaches the bank above ground. The complexity of flow slides requires the use of cutting-edge technological instruments, diving equipment, advanced risk assessment, and a variety of noteworthy probabilistic and sensitivity analyses. Hence, we developed a new sensitivity index to identify the risk of breach failure and vulnerable coastal areas to this risk. In addition, we developed a sophisticated hybrid model that allows for all possibilities of flow slides in sync with random variables used in this new sensitivity index. In this new hybrid model, three distinctive models exist. The 3D Hydrodynamic Model addresses waves, wind, current, climate change, and sediment transport. The Monte Carlo Simulation is responsible for sensitivity analysis, and the Bayesian Network focuses on joint probabilities of coastal flow slide parameters of this new index that incorporates all environmental parameters, including climate change. With the assistance of these three models, researchers aim to: (a) expand the application scope by presenting a method on coastal flow slides; (b) consider different particle diameters corresponding to critical angle slope failure; (c) analyze variables that can play a pivotal role in the flow slides; and (d) present a methodology for coupling coastal flow slide projections with reliable outcomes. The hybrid model incorporates random variables of retrogressive breach failures, and the new risk index considers their ranges to control the simulation. The use of such a hybrid model and risk index offers a robust and computationally efficient approach to evaluating coastal flow slides.

Distributed Sensor Local Linear Fusion Detection of Weak Pulse Signal in Chaotic Background

This paper combines the distributed sensor fusion system with the signal detection under chaotic noise to realize the distributed sensor fusion detection from chaotic background. First, based on the short‐term predictability of the chaotic signal and its sensitivity to small interference, the phase space reconstruction of the observation signal of each sensor is carried out. Second, the distributed sensor local linear autoregressive (DS‐LLAR) model is constructed to obtain the one‐step prediction error of each sensor. Then, we construct a Bayesian risk model and also obtain the corresponding conditional probability density function under each sensor’s hypothesis test which firstly needs to fit the distribution of prediction errors according to the parameter estimation. Finally, the fusion optimization algorithm is designed based on the Bayesian fusion criterion, and the optimal decision rule of each sensor and the optimal fusion rule of the fusion center are jointly solved to effectively detect the weak pulse signal in the observation signal. Simulation experiments show that the proposed method which used a distributed sensor combined with a local linear model can effectively detect weak pulse signals from chaotic background.

State-of-Health Assessment for Aero-Engine Based on Density-Distance Clustering and Fuzzy Bayesian Risk

State-of-health (SOH) assessment for aero-engine can effectively reduce maintenance cost and operational risk, and is also a significant part of the prognostics and health management (PHM) system. However, the current SOH assessment is usually closely coordinated with other parts of PHM to achieve specific functions. This is not conducive to generalizing the function of SOH assessment. Therefore, this paper proposes a data-driven framework of SOH assessment that mainly includes data preprocessing, pseudo label generation, weight assignment and feature selection, and assessment, which enhances the systematicness of SOH assessment. A combination model based on density-distance clustering and fuzzy Bayesian risk models is designed to generate a pseudo label, select optimal parameter subset, and assign weight. Then, two assessment indicators including state membership degree and health degree are produced based on two fuzzy models for horizontal and vertical comparisons. These two indicators expand the dimensions and perspectives of SOH measurement, which can more comprehensively characterize the health state of the engine. Finally, the correctness and effectiveness of the proposed methodology are verified by the widely used Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dataset.

Implementing Bayesian networks for ISO 31000:2018-based maritime oil spill risk management: State-of-art, implementation benefits and challenges, and future research directions

The risk of a large-scale oil spill remains significant in marine environments as international maritime transport continues to grow. The environmental as well as the socio-economic impacts of a large-scale oil spill could be substantial. Oil spill models and modeling tools for Pollution Preparedness and Response (PPR) can support effective risk management. However, there is a lack of integrated approaches that consider oil spill risks comprehensively, learn from all information sources, and treat the system uncertainties in an explicit manner. Recently, the use of the international ISO 31000:2018 risk management framework has been suggested as a suitable basis for supporting oil spill PPR risk management. Bayesian networks (BNs) are graphical models that express uncertainty in a probabilistic form and can thus support decision-making processes when risks are complex and data are scarce. While BNs have increasingly been used for oil spill risk assessment (OSRA) for PPR, no link between the BNs literature and the ISO 31000:2018 framework has previously been made. This study explores how Bayesian risk models can be aligned with the ISO 31000:2018 framework by offering a flexible approach to integrate various sources of probabilistic knowledge. In order to gain insight in the current utilization of BNs for oil spill risk assessment and management (OSRA-BNs) for maritime oil spill preparedness and response, a literature review was performed. The review focused on articles presenting BN models that analyze the occurrence of oil spills, consequence mitigation in terms of offshore and shoreline oil spill response, and impacts of spills on the variables of interest. Based on the results, the study discusses the benefits of applying BNs to the ISO 31000:2018 framework as well as the challenges and further research needs.

Assessment of the impact of geogenic and climatic factors on global risk of urinary stone disease

Urinary Stone Disease (USD) or urolithiasis has plagued humans for centuries, and its prevalence has increased over the past few decades. Although USD pathology could vary significantly among individuals, previous qualitative assessments using limited survey data demonstrated that the prevalence of USD might exhibit a distinctive geographical distribution (the so-called “stone belt”), without any knowledge about the characteristics and contribution factors of the belt. Here, we argue that the spatial distribution of USD can at least partly be explained by geogenic and climatic factors, as it correlates with the ambient geo-environmental conditions modulated by lithology/mineralogy, water quality and climate. Using a Bayesian risk model, we assessed the global risk of USD based on updated big data of four key geogenic factors: phosphorite mines (inventory >1600 points), carbonate rocks (at the scale of 1:40 million), Ca2+/Mg2+ molar ratio of river water (1.27 million samples distributed over 17,000 sampling locations), and mean air temperature (0.5o × 0.5° resolution) representing the climate. We quantitatively identified possible contributions of the factors to USD and delineated the regions with the high USD risk which stretched from southern North America, via the Mediterranean region, northeastern Africa, southern China to Australia, and roughly coincide with the world's major areas of carbonate outcropping. Under current climate conditions, the areas with the probabilities for the USD prevalence of ≥50% and ≥30% covered 3.7% and 20% of the Earth's land surface, respectively. By the end of the 21st century, such total areas could rise to 4.4% and 25% as a result of global warming. Since the USD data used in this study were quite heterogeneous, the prediction results needed further calibration with additional high-quality prevalence data in the future.

Retrospective Evaluation of Bayesian Risk Models of LVAD Mortality at a Single Implant Center

Use of a left ventricular assist device (LVAD) can benefit patients with end stage heart failure, but only with careful patient selection. In this study, previously derived Bayesian network models for predicting LVAD patient mortality at 1, 3, and 12 months post-implant were evaluated on retrospective data from a single implant center. The models performed well at all three time points, with a receiver operating characteristic area under the curve (ROC AUC) of 78, 76, and 75%, respectively. This evaluation of model performance verifies the utility of these models in “real life” scenarios at an individual institution.

Ensemble Risk Model of Emergency Admissions (ERMER)

IntroductionAbout half of hospital readmissions can be avoided with preventive interventions. Developing decision support tools for identification of patients’ emergency readmission risk is an important area of research. Because, it remains unclear how to design features and develop predictive models that can adjust continuously to a fast-changing healthcare system and population characteristics. The objective of this study was to develop a generic ensemble Bayesian risk model of emergency readmission. MethodsWe produced a decision support tool that predicts risk of emergency readmission using England's Hospital Episode Statistics inpatient database. Firstly, we used a framework to develop an optimal set of features. Then, a combination of Bayes Point Machine (BPM) models for different cohorts was considered to create an optimised ensemble model, which is stronger than the individual generative and non-linear classifications. The developed Ensemble Risk Model of Emergency Admissions (ERMER) was trained and tested using three time-frames: 1999–2004, 2000–05 and 2004–09, each of which includes about 20% of patients in England during the trigger year. ResultsComparisons are made for different time-frames, sub-populations, risk cut-offs, risk bands and top risk segments. The precision was 71.6–73.9%, the specificity was 88.3–91.7% and the sensitivity was 42.1–49.2% across different time-frames. Moreover, the Area Under the Curve was 75.9–77.1%. ConclusionsThe decision support tool performed considerably better than the previous modelling approaches, and it was robust and stable with high precision. Moreover, the framework and the Bayesian model allow the model to continuously adjust it to new significant features, different population characteristics and changes in the system.

50th Anniversary Landmark Commentary on Edwards FH, Clark RE, Schwartz M. Coronary artery bypass grafting: The Society of Thoracic Surgeons National Database experience. Ann Thorac Surg 1994;57:12–9

50th Anniversary Landmark Commentary on Edwards FH, Clark RE, Schwartz M. Coronary artery bypass grafting: The Society of Thoracic Surgeons National Database experience. Ann Thorac Surg 1994;57:12–9

Gastric cancer incidence and geographical variations: the influence of gender and rural and socioeconomic factors, Zaragoza (Spain)

Socioeconomic status (SES) and rural factors have been shown to be associated with gastric cancer epidemiology. The aim of this study was to identify geographical variations in gastric cancer incidence in Zaragoza province (Spain) during the period 1993-2002, and their association with SES and rural factors. Incident cases were extracted from the population-based Zaragoza Cancer Registry. The geographical analysis unit was the census tract (CT) in Zaragoza city (N = 462) and the municipalities for the rest of the province (N = 292). Four indexes were applied: two deprivation and two rurality indexes, included in a Bayesian risk model discretized in quartiles. Standardized incidence ratios (SIRs) were calculated using the incidence rates in Spain. SIRs were adjusted by a Bayesian generalized linear mixed model (GLMM). From 1993 to 2002, 1,309 cases of gastric cancer were registered in Zaragoza city and 578 in the rest of the province. High risk was observed in CTs for the peripheral areas of the city. The incidence risk in men was 2 (95 % confidence interval [CI] 1.22-2.98) times higher in the most deprived CTs compared with the least deprived CTs, but no statistically significant differences were found in women. Municipalities with higher risk were observed in the north of the province, but no significant association was found with SES. Regarding the rurality index, a positive trend was observed in women, but it was statistically significant only for the most rural quartile (2.49, 95 % CI 1.07-4.92). Geographical differences in gastric cancer incidence were detected. Although these differences could be partially explained by the deprivation index for men in Zaragoza city, deprivation index cannot explain geographical differences for women. In the rest of the province, the rurality index 1991 could explain, at least for women, geographical differences. It is still necessary to develop a deprivation index suitable for small municipalities.

Impact of internal mammary artery conduits on operative mortality in coronary revascularization

The long-term advantages of internal mammary artery (IMA) conduits in cotonary artery bypass graft (CABG) procedures are widely recognized, but the immediate short-term impact of IMA grafts is not well defined. The purpose of this study was to investigate the influence of IMA conduits on CABG operative mortality (OM). A retrospective study of two groups of patients undergoing isolated CABG was performed. Patients having at least one IMA graft (group 1) were compared with those with only venous conduits (group 2). The patient population was taken from The Society of Thoracic Surgeons National Cardiac Surgery Database, which contains a broad multiinstitutional experience. A total of 38,578 registered patients undergoing isolated CABG from 1987 through 1991 were studied. Of these, 18,614 patients had at least one IMA conduit (group 1), whereas 19,964 had CABG using entirety venous conduits (group 2). The OM for group 1 was 2.0% ( 365 / 18,614 ), whereas the mortality was 4.5% ( 903 / 19,964 ) for group 2 ( p < 0.005). Patient sub-groups were examined to determine if the improved OM associated with IMA grafting was present in these patient subsets. The population was broken down according to age, sex, ejection fraction, extent of coronary disease, and operative priority. For each subset, univariate analysis showed that group 1 OM was significantly less ( p < 0.005) than the OM for group 2. Numerous combinations of these patient parameters were also analyzed. Group 1 patients had a significant ( p < 0.05) improvement in OM in each combination except for patients more than 70 years of age requiring reoperations. Multivariate analysis also demonstrated that the use of an IMA was independently associated with improved patient survival ( p < 0.0025). To consolidate the net effect of numerous patient risk factors, a Bayesian risk model was used to stratify patients into risk quartiles. For the lowest risk quartile (predicted OM <1.14%) there was no significant difference in group 1 versus group 2 patients. For every other risk category, there was a significant improvement in OM associated with IMA use ( p < 0.005). We conclude that in the vast majority of patient subsets, IMA conduits are associated with a statistically significant improvement in OM when compared with use of venous conduits alone.

A note on Bayesian risk models of audit practice*

Abstract. This note identifies some new issues associated with the use of discrete Bayesian risk models in auditing. The major findings are that for sequential revision, discrete Bayesian risk models may need to be used in a different way than suggested by the professional literature, and some modifications are proposed to the CICA (1980) multiple revision risk model in order to make it more general.Résumé. L'auteur recense certaines questions nouvelles associées à l'utilisation des modèles de risques bayesiens discontinus en vérification. L'étude révèle principalement que dans le cas de l'examen séquentiel, les modèles de risque bayesiens discontinus peuvent devoir être utilisés d'une façon différente de celle que suggèrent les ouvrages spécialisés, et l'auteur propose d'apporter certaines modifications au modèle de risque à révisions multiples de l'I.C.C.A. (1980) afin d'en rendre la portée plus générale.