Multiple sources of meter data can lead to inconsistencies and problems.
When creating a customer bill, calculations must produce matching results regardless of meter data sources.
Customers need to see a consistent data view of their bill.
Often, meter communication issues create data issues and these lead to billing errors.
Fractal built a system for a major electric utility as a parallel application with the system-of-record in less than 90 days.
All interval data was reconciled against daily data, then reconciled against the billing system data.
A machine learning system was built to identify unusual consumption patterns.
These could indicate electrical equipment which was close to end-of-life.
This system was delivered for less than the maintenance cost of the current legacy system and saved the utility $10.2 million for a legacy system rewrite.
This system was built and in production in less than 90 days.