Statistical Analysis of Animal-Related Outages in Power Distribution Systems - A Case Study

This paper presents a data-driven approach for exploring animal-related outages in power distribution systems. The main objectives are to uncover the underlying structure of animal-related outages, identify variables that strongly influence their frequency, and build a model to predict their occurrence on a weekly basis. To carry out this study, we utilize actual outage data obtained from a major utility company in the U.S., in addition to different types of wildlife and weather-related data. Results of this study could be very informative to utility companies in maximizing insight into their animal-related outage problems, and to build parsimonious models for predicting them.