No-Code AI for Critical Infrastructure

Electricity meters are mission-critical assets on Con Edison’s electric grid. The assets are customer-facing and represent safety risks if connection point temperatures increase to dangerous levels. Additionally, overheating causes expensive maintenance and erodes company margins.

Identifying which meters were at-risk was difficult for business analysts. Temperature readings resided in objects on the C3 AI Platform. Filtering point measurements that resided in C3 AI objects to the appropriate electricity meters and time-range required highly technical expertise. Additionally, the data needed to be manipulated to generate a report (e.g., filter the dataset to remove obsolete meters and sort the dataset to enable prioritization of the hottest assets). Connecting to, preparing, and analyzing the data to generate a report required hours of multiple analysts’ time.

Con Edison looked to C3 AI Ex Machina to reduce reporting timelines. In a matter of hours, the energy company’s analysts collaborated to deploy a C3 AI Ex Machina workflow that reduced the reporting cycle to minutes. The workflow is pre-configured with the appropriate datastores, prepares and filters the data, and implements output nodes that allows analysts to visualize at-risk assets and report insights broadly.

For example, the workflow enriches electric meter data with latitude and longitude fields to enable a geospatial visualization, unlocking significant productivity gains. The visualization – depicted below – plots at-risk meters on a map. Any business analyst or field operator with appropriate credentials can view this visualization. Additionally, when business analysts periodically run the C3 Ex Machina workflow, the visualization is automatically generated and sent broadly to field operators to expedite maintenance, thereby enabling collaboration and enhancing productivity and safety.

In future versions of the workflow, business analysts may choose to further integrate the pipeline’s insights into operational processes. Business analysts can write the results of the analysis to an enterprise system – such as the C3 AI Platform, a REST integration service, Salesforce, and more – and expose the results in a production application.

• Connected to enterprise datastores containing temperature readings on 100,000 meters across the United States
• Rapidly prepared data and filtered to the most recent 700,000 meter temperature readings
• Enriched dataset by joining meter dataset with external latitude and longitude dataset
• Added configurations to enhance the flexibility of the reporting workflow, such as time-window of interest and the temperature threshold above which assets are at-risk
• Utilized C3 AI Ex Machina geospatial visualization to quickly identify at-risk meters; operationalized insights by sharing visualizations with the field
• Enhanced analyst productivity by 80%