A framework to assess natural hazard-induced service inoperability in the electricity sector

Our built environment is threatened with ever-increasing risks of climate change and natural hazard- induced extreme events. Recent natural catastrophes especially extreme hydro-climatological hazards such as floods, forest fires, droughts, and heatwaves often times lead to cascading severe weather-induced power outages. These outages affect immense economic loss. In this research, we propose a strategic planning and decision-making framework to predict the inoperability of the utility in face of several types of natural disasters. Inoperability can indicate the level of system malfunction caused by the severe weathers. In this research, inoperability refers to the interruption in power supply, measured by the extents of peak demand due to power outage. The proposed framework consists of the three main phases: (1) natural hazard-induced major disaster selection, (2) relevant feature selection, and (3) identify the factors that influence the levels of inoperability in the utility sector due to power outage (i.e., lack of power supply). First, we propose to leverage Principal Component Analysis (PCA) for dimension reduction and feature selection; then, using the selected features we propose to develop advanced statistical learning models to identify the key factors that will increase the risk of inoperability in the electric power sector. In this paper, we establish our framework using generalized additive model (GAM), although the framework can be easily extended to include other models as well. The data on major power outage events in the continental U.S., ranging between 2000 and 2016, is available to validate our propose framework at the state level. Our proposed framework will allow the decision makers and stakeholders such as the utility companies and state regulatory commissions to understand the risks of inoperability of the regional electricity sectors, help minimize the risk of natural hazard-induced extreme outages in the electricity sector, and improve the security of the electricity sector as a whole.