These are case studies members of our team have taken part in to help achieve our main goal. If you want to access the full study, the link to it is on the last page of each PDF.
Model and solution method for mean-risk cost-based post-disruption restoration of interdependent critical infrastructure networks
by Basem A. Alkhaleel, Haitao Liao, Kelly M. Sullivan
This case study is all about improving the resilience of our critical infrastructure networks and lowering repair costs. The methods included here are an extension of previous work in an attempt to include multiple networks.
Casual Inference in Longitudinal Studies Using Casual Bayesian Network with Latent Variables
by Phat Huynh, Leah Irish, Arveity Setty, Om Yadav, Trung Q. Le
This case study expands on Bayesian Networks by addressing the limitations of the Casual Bayesian Network (CBN) with the use of a Bayesian Network for Latent-Variable framework.
Condition-Based Monitoring as a Robust Strategy Towards Sustainable and Resilient Multi-Energy Infrastructure Systems
by Nita Yodo, Tanzina Afrin, Om Prakash Yadav, Di Wu, Ying Huang
This case study aims to mitigate the impact of disturbances of energy supply by making the infrastructure resilient. By using condition-based monitoring (CBM) and aggregating data of the systems, we can monitor when and where maintenance is needed.
Sustainable Development for Oil and Gas Infrastructure from Risk, Reliability, and Resilience Perspectives
by Yasir Mahmood, Tanzina Afrin, Ying Huang, and Nita Yodo
This case study focuses on reaching sustainability for the oil and gas industry by looking at two different matrices, the Fundamental and Coupling matrices. These matrices allow us to better understand direct and indirect impacts toward the goal of sustainable development.
Convolutional Non-homogeneous Poisson Process with Application to Wildfire Risk Quantification for Power Delivery Networks
by Guanzhou Wei, Xiao Liu, and Feng Qiu
This case study elaborates on how certain power lines are susceptible to wildfires and how to predict the risks associated with it by using a new spatio-temporal point process model on a linear network.
Predicting Natural Gas Pipeline Failures Caused by Natural Forces: An Artificial Intelligence Classification Approach
by Bright Awuku, Ying Huang, and Nita Yodo
This case study devises an AI algorithm to predict damage from natural forces to natural gas pipelines with additional climate change information for the most accurate forecasts. This information can be used for things like locating problem areas and improving safety while lowering costs.
A Physics-informed Latent Variables of Corrosion Growth in Oil and Gas Pipelines
by Phat K. Huynh, Abdulsalam A. Alqarni, Om P. Yadav,
and Trung Q. Le
This case study composes a more physics-informed model instead of deterministic or stochastic-process-based models that are currently used to predict corrosion growth in pipelines, and introduces previously unaddressed physics-related variables.