How Central is putting AI to work
Artificial intelligence is changing how organizations find information, analyze data and approach everyday work. At Central Electric Power Cooperative, employees are exploring ways to use these tools to better serve our member cooperatives.
Two applications are already in use. One helps employees identify unusual patterns in meter data before billing, while another makes it easier to search service contracts. Central staff monitors both tools and uses the findings to make decisions.
Supporting the billing process
Accurate meter data helps Central ensure its member cooperatives are billed correctly for the power they receive. Before billing its member cooperatives, Central reviews data from hundreds of meters across its system. Examining this data can be time-consuming, and unusual patterns aren’t always obvious or easy to catch.
Central developed a machine learning model that helps with this process by comparing a meter’s current load curve with its typical pattern. The model uses recent meter data and weather information to flag anomalies that may need further review.
Central’s Billing and Engineering team then investigates. A flagged pattern could point to missing data, a meter configuration issue or a change in electricity use.
Not every flagged pattern indicates a problem. Some reflect expected changes in electricity use, such as increased demand during unusually hot or cold weather or a change in a facility’s operating schedule.
Still, the model helps double-check Central’s existing billing validation process. And the model will only get better and better as staff continue to train and refine it.
Finding information in contracts
Central maintains approximately 200 service contracts. Finding a specific provision can require employees to search through and compare multiple documents.
Central developed an AI-powered search tool to help authorized employees find information within those contracts. Employees can ask questions in everyday language, and the tool returns relevant passages and summaries with details identifying the contract, page and section. Employees can then review the original language.
Building the tool also provided lessons for Central’s team. Early in development, searches based on similar language did not always identify the specific contract an employee needed. The team refined the search capabilities to improve how the tool finds specific content.
Technology supported by employee expertise
For Central, these applications offer ways to make existing work more efficient while keeping employee expertise at the center of the process. The tools help staff focus their attention, verify information and continue providing reliable service to member cooperatives.
As Central gains experience with AI, employee feedback will help guide how these tools are refined and used.
