Real-time risk analysis with optimization proxies Chen, Wenbo; Tanneau, Mathieu; Van Hentenryck, Pascal
Electric power systems research,
October 2024, 2024-10-00, 2024-10-01, Volume:
235, Issue:
C
Journal Article
Peer reviewed
The increasing penetration of renewable generation and distributed energy resources requires new operating practices for power systems, wherein risk is explicitly quantified and managed. However, ...traditional risk-assessment frameworks are not fast enough for real-time operations, because they require numerous simulations, each of which requires solving multiple economic dispatch problems sequentially. The paper addresses this computational challenge by proposing proxy-based risk assessment, wherein optimization proxies are trained to learn the input-to-output mapping of an economic dispatch optimization solver. Once trained, the proxies make predictions in milliseconds, thereby enabling real-time risk assessment. The paper leverages self-supervised learning and end-to-end-feasible architecture to achieve high-quality sequential predictions. Numerical experiments on large systems demonstrate the scalability and accuracy of the proposed approach.
•Rising renewables and distributed resources need new power system risk assessments.•A new risk assessment framework utilizing fast optimization proxies is proposed.•The framework uses an end-to-end feasible proxy for scalable and accurate training.•Proposed risk assessment is highly accurate and 30x faster than optimization methods.
Regulating chemical mixtures is a complex scientific and policy task. The aim of this study was to investigate typical mixtures and their potential risks based on internal exposure levels in the ...European population. Based on human biomonitoring (HBM) data made available via the HBM4EU project, we derived generic mixtures representative of a median (P50) and a worst-case scenario (P95) for adults and children. We performed a mixture risk assessment based on HBM concentrations, health-based guidance values (HBGVs) as internal thresholds of concern, and the conservative assumption of concentration addition applied across different toxicological endpoints. Maximum cumulative ratios (MCRs) were calculated to characterize the mixture risk. The mixtures comprise 136 biomarkers for adults and 84 for children, although concentration levels could be quantified only for a fraction of these. Due to limited availability of HBGVs, the mixture risk was assessed for a subset of 20 substance-biomarker pairs for adults and 17 for children. The mixture hazard index ranged from 2.8 (P50, children) to 9.2 (P95, adults). Six to seven substances contributed to over 95% of the total risk. MCR values ranged between 2.6 and 5.5, which is in a similar range as in previous studies based on human external exposures assessments. The limited coverage of substances included in the calculations and the application of a hazard index across toxicological endpoints argue for caution in the interpretation of the results. Nonetheless the analyses of MCR and MAF
can help inform a possible mixture assessment factor (MAF) applicable to single substance risk assessment to account for exposure to unintentional mixtures.
Objective To compare the predictive accuracy and clinical utility of five risk scoring systems in the assessment of patients with upper gastrointestinal bleeding.Design International multicentre ...prospective study.Setting Six large hospitals in Europe, North America, Asia, and Oceania.Participants 3012 consecutive patients presenting over 12 months with upper gastrointestinal bleeding.Main outcome measures Comparison of pre-endoscopy scores (admission Rockall, AIMS65, and Glasgow Blatchford) and post-endoscopy scores (full Rockall and PNED) for their ability to predict predefined clinical endpoints: a composite endpoint (transfusion, endoscopic treatment, interventional radiology, surgery, or 30 day mortality), endoscopic treatment, 30 day mortality, rebleeding, and length of hospital stay. Optimum score thresholds to identify low risk and high risk patients were determined.Results The Glasgow Blatchford score was best (area under the receiver operating characteristic curve (AUROC) 0.86) at predicting intervention or death compared with the full Rockall score (0.70), PNED score (0.69), admission Rockall score (0.66, and AIMS65 score (0.68) (all P<0.001). A Glasgow Blatchford score of ≤1 was the optimum threshold to predict survival without intervention (sensitivity 98.6%, specificity 34.6%). The Glasgow Blatchford score was better at predicting endoscopic treatment (AUROC 0.75) than the AIMS65 (0.62) and admission Rockall scores (0.61) (both P<0.001). A Glasgow Blatchford score of ≥7 was the optimum threshold to predict endoscopic treatment (sensitivity 80%, specificity 57%). The PNED (AUROC 0.77) and AIMS65 scores (0.77) were best at predicting mortality, with both superior to admission Rockall score (0.72) and Glasgow Blatchford score (0.64; P<0.001). Score thresholds of ≥4 for PNED, ≥2 for AIMS65, ≥4 for admission Rockall, and ≥5 for full Rockall were optimal at predicting death, with sensitivities of 65.8-78.6% and specificities of 65.0-65.3%. No score was helpful at predicting rebleeding or length of stay.Conclusions The Glasgow Blatchford score has high accuracy at predicting need for hospital based intervention or death. Scores of ≤1 appear the optimum threshold for directing patients to outpatient management. AUROCs of scores for the other endpoints are less than 0.80, therefore their clinical utility for these outcomes seems to be limited.Trial registration Current Controlled Trials ISRCTN16235737.
Risk assessment (RA) and life cycle assessment (LCA) are two analytical tools used to support decision making in environmental management. This study reviewed 30 environmental assessment case studies ...that claimed an integration, combination, hybridization, or complementary use of RA and LCA. The focus of the analysis was on how the respective case studies evaluated emissions of chemical pollutants and pathogens. The analysis revealed three clusters of similar case studies. Yet, there seemed to be little consensus as to what should be referred to as RA and LCA, and when to speak of combination, integration, hybridization, or complementary use of RA and LCA. This paper provides clear recommendations toward a more stringent and consistent use of terminology. Blending elements of RA and LCA offers multifaceted opportunities to adapt a given environmental assessment case study to a specific decision making context, but also requires awareness of several implications and potential pitfalls, of which six are discussed in this paper. To facilitate a better understanding and more transparent communication of the nature of a given case study, this paper proposes a “design space” (i.e., identification framework) for environmental assessment case studies blending elements of RA and LCA. Thinking in terms of a common design space, we postulate, can increase clarity and transparency when communicating the design and results of a given assessment together with its potential strengths and weaknesses.
Millions of people are exposed to arsenic resulting in a range of health implications. This paper provides an up-to-date review of the different sources of arsenic (water, soil, and food), indicators ...of human exposure (biomarker assessment of hair, nail, urine, and blood), epidemiological and toxicological studies on carcinogenic and noncarcinogenic health outcomes, and risk assessment approaches. The review demonstrates a need for more work evaluating the risks of different arsenic species such as, arsenate, arsenite monomethylarsonic acid, monomethylarsonous acid, dimethylarsinic acid, and dimethylarsinous acid as well as a need to better integrate the different exposure sources in risk assessments.
Objective
Empirically informed suicide risk assessment frameworks are useful in guiding the evaluation and treatment of individuals presenting with suicidal symptoms. Joiner et al. (1999) formulated ...one such framework, which has provided a concise heuristic for the assessment of suicide risk. The purpose of this review is to ensure compatibility of this suicide risk assessment framework with the growing literature on suicide‐related behaviors.
Methods
This review integrates recent literature on suicide risk factors and clinical applications into the existing model. Further, we present a review of risk factors not previously included in the Joiner et al. (1999) framework, such as the interpersonal theory of suicide variables of perceived burdensomeness, thwarted belongingness, and capability for suicide (Joiner, 2005; Van Orden et al., 2010) and acute symptoms of suicidality (i.e., agitation, irritability, weight loss, sleep disturbances, severe affective states, and social withdrawal).
Results
These additional indicators of suicide risk further facilitate the classification of patients into standardized categories of suicide risk severity and the critical clinical decision making needed for the management of such risk.
Conclusions
To increase the accessibility of empirically informed risk assessment protocols for suicide prevention and treatment, an updated suicide risk assessment form and decision tree are provided.