Abstract: Severe weather events lead to widespread, prolonged power outages, largely in the power distribution system. Utilities face significant resource allocation challenges in preparing for and responding to forecast weather events. How many external crews should be requested? How much extra material should be stockpiled? Where should it be located? These decisions must be made ahead of the weather event under considerable uncertainty about the impacts of an uncertain weather forecast. Being able to forecast weather impacts prior to a weather event can be of substantial benefit to electric power utilities. In this talk I will present models developed over the past 14 years that aim to forecast power outages due to weather events. The types of events addressed span hurricanes, thunderstorms, heat events, and winter events. Both statistically-based and engineering fragility function basedapproaches will be discussed in the talk. I will close with a discussion of the challenges associated with operational use of these models for severe weather events.
About: Seth Guikemais a Professor in the Department of Industrial and Operations Engineering and the Department Civil and Environmental Engineering at the University of Michigan. He is also a Professor II in the Department of Economics, Planning, and Safety at theUniversity of Stavanger (Norway) and Data Science Research Follow at One Concern, Inc., a Silicon Valley based startup focused on data analytics for community resilience. He is the President-Elect of the Society for Risk Analysis and the Area Editor for the Mathematical Modeling area of Risk Analysis.