2027 Event Agenda

At Duquesne Light Company (DLC), serving more than 600,000 customers across an 812 square‑mile territory in Pennsylvania, we reached a point where grid planning work was being slowed by manual data extraction and spreadsheet analysis. As AMI data volumes grew, it became harder to repeat studies consistently and validate connectivity models fast enough to support planning and operational needs.

We changed our approach by using AMI interval data to generate load profiles and validate connectivity at multiple levels—service point, neighborhood, and system. We were deliberate in how we introduced the change. We used a phased “crawl, walk, run” rollout so we could validate early results manually, compare findings against methods our teams already trusted, and align across planning, GIS, and operations before expanding automation into routine workflows.

The impact showed up in weekly planning work. Tasks that used to take hours in spreadsheets can now be completed in minutes with repeatable steps. We improved the consistency of load curve and trend analysis, reduced time spent chasing data issues and gained clearer visibility into load behavior that supports planning decisions and targeted investigations. This session shares the practical lessons we learned along the way, including what worked, what didn’t, and what we’d do differently.

Attendees will learn how we:

·         Reduced spreadsheet-heavy grid planning work by moving to repeatable, AMI-driven analysis workflows

·         Used AMI interval data to validate connectivity and build load profiles at the service point, neighborhood, and system levels

·         Built trust through a phased rollout with manual validation first, then automation as cross-team confidence grew

·         Established a feedback loop between analysis results and grid model updates to improve accuracy over time