It is moderately concentrated at the top and long-tailed below it. The five leading assignees together hold 124 of the 312 records in scope, or 39.7%, and the top ten extend that to 172 records, or 55.1%. That leaves close to half the field spread across 90 or more additional filers, many with just one or two records, so freedom-to-operate risk is not evenly distributed — it clusters around a small group of repeat filers plus a much larger set of isolated filings.

Filings rose steadily from 8 in 2017 to a peak of 29 in 2021, then fell to 22 by 2024, a -24% change over that three-year span. Because publication lags filing by roughly 18 months, 2025 and 2026 figures in any dataset are still incomplete and should not be read as evidence of continued decline. The honest read is that the field expanded quickly around 2019-2021 and has since settled to a lower but still active filing rate.

G06Q, the business-and-admin data-processing class, appears on 63.8% of the 312 records in this dataset, ahead of G06N for AI-based computing at 23.4% and G06F for general electric data processing at 22.1%. Power-system-specific classes are present but secondary: H02J covers 11.9% of records and control classes G05B and G05D cover under 5% each. This mix indicates that much of the claim language is written as generalized optimization or risk-computation method claims rather than power-hardware or grid-control claims, which matters for anyone trying to scope a search or a design-around.

The ranked dataset covers 100 companies and individual filers across 312 records, with a single leading assignee holding 38 records compared with 16 for the fifth-ranked name and 9 for the tenth. Several of the most active named assignees, however, show zero filings in the most recent tracked year, which given the 18-month publication lag may reflect pending filings still working through the pipeline rather than a genuine exit from the field. Reviewing recent-year momentum alongside total record count gives a fuller picture than ranking alone.

US20110276514A1, assigned to International Business Machines Corporation and published in 2011, describes a computing system that evaluates the risk-robustness of a utility's capacity resource plan by computing a net present value of total cost distribution, calculating a standardized risk measure over that value, and then optimizing the resource plan to minimize that measure. It is one of the earlier records in this corpus and anchors the risk-measure-over-NPV framing seen in later filings, but whether it blocks a new filing depends entirely on how narrowly or broadly its granted claims define the risk measure and the optimization step — that requires reading the claim set directly, not the abstract.