Citizen Portal
Sign In

Get Full Government Meeting Transcripts, Videos, & Alerts Forever!

Get email alerts on the Load Forecasting topic

No spam. Unsubscribe anytime.

FERC panel hears experts on forecasting challenges posed by large data‑center and industrial loads

Federal Energy Regulatory Commission · July 9, 2026
AI-Generated Content: All content on this page was generated by AI to highlight key points from the meeting. For complete details and context, we recommend watching the full video. so we can fix them.

Summary

FERC staff convened industry, regulator and customer representatives to discuss how new 'large loads' (notably data centers) should be represented in short‑term operations and long‑range transmission planning. Panelists urged clearer data sharing, confidence/maturity metrics and scenario‑based forecasts to reduce costly over‑ or under‑investment.

Federal Energy Regulatory Commission staff convened a public panel on load forecasting in docket 8010‑12‑17 on July 13, 2026, bringing together RTO and ISO forecasters, NERC, consumer advocates, consultants and industry customers to discuss how large new loads should be integrated into planning and operations. Chris Young, a division director in FERC’s Office of Technical Reporting and Economics, opened the workshop by saying load forecasts are “a critical input” for resource adequacy, transmission planning and reliable operation of the bulk power system.

The panel focused on two linked problems: the technical challenge of incorporating large, rapidly deployed loads into statistical and engineering forecasts, and the procedural challenge of getting reliable, verifiable information early enough to guide multi‑billion‑dollar planning decisions. D’L Oates, executive director of Markets and Grid Research at Midcontinent ISO, said MISO expects “20 to 30 gigawatts of new large load by 2030,” and described a 20‑year, scenario‑based forecast MISO uses for long‑range transmission planning. John Mora of the North American Electric Reliability Corporation said that while many regions’ 10‑year forecasts have historically tracked within roughly 10% of realized demand, the recent wave of large computational and industrial loads makes it more important to “characterize a range of plausible futures” rather than rely on a single point forecast.

Consumer advocates and market observers urged stronger accountability for who owns and certifies load projections. Greg Polis, executive director of the consumer advocates of the PJM states, said forecasts in the 2010–2018 period tended to overstate demand, a bias that can drive unnecessary infrastructure costs ultimately borne by customers. Polis and other panelists recommended clear confidence or maturity tiers—high, medium, low—backed by objective milestones such as executed energy supply agreements, financial commitments, or construction progress so planners know which loads are likely to materialize.

Industry participants described the data and operational detail they can provide, and where gaps remain. Mudita Suri of Amazon Web Services said AWS shares nameplate capacity, ramp profiles and, where appropriate, hourly load shapes with utility partners and supports standardized definitions so planners “speak the same language.” She also called for clarity on what commitments qualify a proposed load for inclusion in base‑case planning. Several panelists said proprietary datasets and nondisclosure agreements currently prevent regulators and some stakeholders from seeing the inputs behind forecasts, complicating validation and cross‑region comparison.

Panelists recommended practical steps to reduce uncertainty: (1) adopt common terminology and a maturity/confidence assessment so forecasts are comparable across regions; (2) collect more granular operational data (ramp rates, hourly profiles, behind‑the‑meter arrangements) under defined confidentiality protections; (3) use scenario or probabilistic forecasts and confidence bands for long‑range transmission planning; and (4) improve post‑hoc validation so forecasters can measure what inputs were most predictive.

The Energy Systems Integration Group report on large‑load forecasting was highlighted as a source of near‑term recommendations and longer‑term institutional actions; John Wilson, who led that project team, said the group offered seven short‑term recommendations and three longer‑term research and institutional priorities. Panelists also pointed to ongoing NERC workstreams and EPRI efforts to facilitate data sharing and industry standards.

FERC staff said they will issue a notice inviting comments in the docket (8010‑12‑17) and make a workshop recording and transcript available. Commissioner Lacerte closed by urging continued collaboration and faster progress, noting congressional attention to forecasting issues.

The workshop produced no formal votes or binding actions; it concluded with staff planning to solicit written comments in the docket and continue engagement with RTOs, NERC and stakeholders.