Predictive approaches to prevent outages, control costs and serve customers
Statistical models guide where and when an electrical utility trims vegetation
A major cost for electricity suppliers is managing tree growth to keep transmission “feeder” lines clear of overgrowth and prevent outages. A southern electric utility with operations in three states spends up to $100 million a year on this effort. The company wanted to explore ways to both defend this budget in the face of cost-cutting measures and optimize vegetation management.
NTT DATA and its Decision Architecture framework enabled the utility to move up the analytics maturity curve to help the company understand not only what happened, but why it happened, what will happen and how it can influence future decision-making.
Besoin de l’entreprise
A major cost for electricity suppliers is managing tree growth to keep transmission feeder lines clear of overgrowth and prevent outages. This utility wanted to defend its annual $100 million budget to avoid cost-cutting measures and optimize its vegetation management practices.
The company also wanted to test a hypothesis of whether using statistical models to guide where and when to trim vegetation around feeder lines could increase reliability and reduce costs by avoiding outages.
Résultats
- Experiences fewer outages related to overgrown vegetation
- Attains higher levels of continuous service
By giving the utility the insight to trim the right vegetation at the right time, this predictive solution is estimated to save $20 million to $30 million a year in outage-related costs.
Solution
The utility’s current rotating schedules for trimming were calendar based and didn’t account for financial impacts of outages. Nor did they consider a key metric: customer minutes interrupted.
Through the Decision Architecture process, the team from NTT DATA developed a statistical model to rank feeders in order of trim priority based on projected CMI savings, not just on a time-based schedule. A series of Microsoft Power BI visualizations, shared across the three operating companies, included integrated predictive models to support business analysis and boost decision-making power.
By giving the utility the insight to trim the right vegetation at the right time, this predictive solution is estimated to save $20 million to $30 million a year in outage-related costs. It also enables higher levels of continuous service to customers by reducing sustained interruptions due to overgrown vegetation.
About the case study
An electric utility with operations in three U.S. states leverages a decision-driven analytics framework and Microsoft Power BI visualizations to improve business decision-making.