Analysis of hourly weather forecasts as an indicator of outage characteristics in an electric power service system

Open Access
Ananthakrishnan, Vivek
Graduate Program:
Industrial Engineering
Master of Science
Document Type:
Master Thesis
Date of Defense:
Committee Members:
  • Vittaldas V Prabhu, Thesis Advisor
  • Electric Power
  • Outage
  • Predictive Analytics
  • Weather
The world is so dependent on electricity in the 21st century that any interruption or failure in the electric service system could lead to major disruption of daily life and economic losses. Therefore electric utility companies strive to maintain high customer service levels and try to make the best use of past power outage experiences. Increasing capabilities of analytics and improved accuracy of short-term weather forecasts present opportunities for electric utilities to leverage large amount of historic data to forecast outages and proactively make decisions to minimize disruptions. This thesis presents an analytic approach to evaluate the possibility of using hourly weather forecasts to predict outage impact characteristics such as customer minutes interrupted (CMI) and duration of an outage. The key variables in predicting outage duration are identified and their effect on predicting the characteristics is estimated. Furthermore, a cause-effect analysis is done to determine the equipment affected due to various causes of outage, which would help reduce time to repair and enable companies to develop plans to ensure better allocation of resources during an outage and prioritize maintenance schedules in advance of an event. Data from a utility company service location is used as a case study to discuss the methodology proposed and validate the models.