Introduction

---

This project analyzes major power outage events in the continental United States from January 2000 to July 2016, investigating which factors affect the duration and intensity of power outages, and whether we can preemptively predict and detect large-scale power outages.

The dataset contains 1,534 rows representing individual power outage events. The key columns relevant to this analysis include:

• OUTAGE.DURATION (minutes): Duration of the power outage
• CUSTOMERS.AFFECTED: Number of customers impacted by the outage
• CAUSE.CATEGORY: Primary cause of the outage (severe weather, equipment failure, etc.)
• U.S._STATE: State where the outage occurred
• CLIMATE.CATEGORY: Climate conditions during the outage (normal, cold, warm)
• POPULATION: State population
• POPPCT_URBAN (%): Percentage of state population in urban areas
• NERC.REGION: North American Electric Reliability Corporation region

---