Natural gas pipelines are susceptible to external and internal risk factors, such as corrosion, environmental conditions, external interferences, construction and design faults, and equipment failures. Bayesian Networks (BN) is a promising risk assessment approach widely used to evaluate these risk factors. One of BN's inherent limitations is its inability to accurately capture statistical dependencies and causal relationships, which can be overcome by incorporating expert elicitation into BN. To account for uncertainty and vagueness in assessing pipeline failure risks, fuzzy set theory (FST) can be combined with BN, commonly known as Fuzzy Bayesian Networks (FBN). This study developed an FBN framework that uses linguistic variables to calculate fuzzy probability (FPr) through domain expert elicitation, and crisp probabilities (CPr) are computed using historical incident data from the Pipeline and Hazardous Materials Safety Administration (PHMSA). Based on the findings from the case study of the Midwest region of the USA, external factors, i.e., third-party interference, outside force, and other incidents, significantly impact pipeline performance and reliability. Diagnosis inference indicates that in the Midwest region of the USA, pipeline material and age are critical factors leading to corrosion failure by threatening pipeline integrity. The findings from this study suggested that a targeted risk mitigation strategy is paramount for minimizing the risks associated with pipeline networks.
more »
« less
This content will become publicly available on November 1, 2025
Enhancing Risk Assessment in Natural Gas Pipelines Using a Fuzzy Aggregation Approach Supported by Expert Elicitation
Although the natural gas pipeline network is the most efficient and secure transportation mode for natural gas, it remains susceptible to external and internal risk factors. It is vital to address the associated risk factors such as corrosion, third-party interference, natural disasters, and equipment faults, which may lead to pipeline leakage or failure. The conventional quantitative risk assessment techniques require massive historical failure data that are sometimes unavailable or vague. Experts or researchers in the same field can always provide insights into the latest failure assessment picture. In this paper, fuzzy set theory is employed by obtaining expert elicitation through linguistic variables to obtain the failure probability of the top event (pipeline failure). By applying a combination of T- and S-Norms, the fuzzy aggregation approach can enable the most conservative risk failure assessment. The findings from this study showed that internal factors, including material faults and operational errors, significantly impact the pipeline failure integrity. Future directions should include sensitivity analyses to address the uncertainty in data to ensure the reliability of assessment results.
more »
« less
- Award ID(s):
- 1750316
- PAR ID:
- 10522977
- Publisher / Repository:
- ASCE Library
- Date Published:
- Journal Name:
- Practice Periodical on Structural Design and Construction
- Volume:
- 29
- Issue:
- 4
- ISSN:
- 1084-0680
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Although electricity transmission systems are typically very robust, the impacts that arise when they are disrupted motivate methods for analyzing outage risk. For example, N-k interdiction models were developed to characterize disruptions by identifying the sets of k power system components whose failure results in “worst case” outages. While such models have advanced considerably, they generally neglect how failures outside the power system can cause large-scale outages. Specifically, failures in natural gas pipeline networks that provide fuel for gas-fired generators can affect the function of the power grid. In this study, we extend N-k interdiction modeling to gas pipeline networks. We use recently developed convex relaxations for natural gas flow equations to yield tractable formulations for identifying sets of k components whose failure can cause curtailment of natural gas delivery. We then present a novel cutting-plane algorithm to solve these problems. Finally, we use test instances to analyze the performance of the approach in conjunction with simulations of outage effects on electrical power grids.more » « less
-
Existing natural gas pipelines can facilitate low-cost, large-scale hydrogen transportation and storage, but hydrogen may entail safety challenges. These challenges stem from hydrogen’s different properties compared to natural gas, such as higher ignition probability, different flame behavior, and potential for hydrogen embrittlement. Although risk assessments for hydrogen pipelines are increasing, the impact of hydrogen on the risk of third-party excavation damage (TPD), the major cause of pipeline incidents in the U.S., has received little attention. This work presents the SHyTERP model for Safe Hydrogen Transportation and Excavation Risk Prevention for Pipelines. The model incorporates causal models, excavation damage and pipeline failure statistics, and validated physical models of hydrogen and natural gas release and jet flame behavior. Through four case studies, the model compares the TPD risks of hydrogen and natural gas pipelines, offering insights and recommendations for the safe implementation of hydrogen in existing pipelines.more » « less
-
Pipeline networks are a crucial component of energy infrastructure, and natural force damage is an inevitable and unpredictable cause of pipeline failures. Such incidents can result in catastrophic losses, including harm to operators, communities, and the environment. Understanding the causes and impact of these failures is critical to preventing future incidents. This study investigates artificial intelligence (AI) algorithms to predict natural gas pipeline failures caused by natural forces, using climate change data that are incorporated into pipeline incident data. The AI algorithms were applied to the publicly available Pipeline and Hazardous Material Safety Administration (PHMSA) dataset from 2010 to 2022 for predicting future patterns. After data pre-processing and feature selection, the proposed model achieved a high prediction accuracy of 92.3% for natural gas pipeline damage caused by natural forces. The AI models can help identify high-risk pipelines and prioritize inspection and maintenance activities, leading to cost savings and improved safety. The predictive capabilities of the models can be leveraged by transportation agencies responsible for pipeline management to prevent pipeline damage, reduce environmental damage, and effectively allocate resources. This study highlights the potential of machine learning techniques in predicting pipeline damage caused by natural forces and underscores the need for further research to enhance our understanding of the complex interactions between climate change and pipeline infrastructure monitoring and maintenance.more » « less
-
Background Heart failure is a leading cause of mortality and morbidity worldwide. Acute heart failure, broadly defined as rapid onset of new or worsening signs and symptoms of heart failure, often requires hospitalization and admission to the intensive care unit (ICU). This acute condition is highly heterogeneous and less well-understood as compared to chronic heart failure. The ICU, through detailed and continuously monitored patient data, provides an opportunity to retrospectively analyze decompensation and heart failure to evaluate physiological states and patient outcomes. Objective The goal of this study is to examine the prevalence of cardiovascular risk factors among those admitted to ICUs and to evaluate combinations of clinical features that are predictive of decompensation events, such as the onset of acute heart failure, using machine learning techniques. To accomplish this objective, we leveraged tele-ICU data from over 200 hospitals across the United States. Methods We evaluated the feasibility of predicting decompensation soon after ICU admission for 26,534 patients admitted without a history of heart failure with specific heart failure risk factors (ie, coronary artery disease, hypertension, and myocardial infarction) and 96,350 patients admitted without risk factors using remotely monitored laboratory, vital signs, and discrete physiological measurements. Multivariate logistic regression and random forest models were applied to predict decompensation and highlight important features from combinations of model inputs from dissimilar data. Results The most prevalent risk factor in our data set was hypertension, although most patients diagnosed with heart failure were admitted to the ICU without a risk factor. The highest heart failure prediction accuracy was 0.951, and the highest area under the receiver operating characteristic curve was 0.9503 with random forest and combined vital signs, laboratory values, and discrete physiological measurements. Random forest feature importance also highlighted combinations of several discrete physiological features and laboratory measures as most indicative of decompensation. Timeline analysis of aggregate vital signs revealed a point of diminishing returns where additional vital signs data did not continue to improve results. Conclusions Heart failure risk factors are common in tele-ICU data, although most patients that are diagnosed with heart failure later in an ICU stay presented without risk factors making a prediction of decompensation critical. Decompensation was predicted with reasonable accuracy using tele-ICU data, and optimal data extraction for time series vital signs data was identified near a 200-minute window size. Overall, results suggest combinations of laboratory measurements and vital signs are viable for early and continuous prediction of patient decompensation.more » « less
