Bearing RUL Prediction Patents: Leaders, Trends & White Space 2026
Filing growth compares 2021 (261 records) with 2024 (252) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 3,086 records in scope (CR5), not by the ranked leaders only.
What this landscape covers
This dataset tracks patent families that combine bearing remaining-useful-life (RUL) prediction — RUL modelling, bearing life prediction, prognostic bearing models — with predictive maintenance infrastructure: condition monitoring, equipment fault diagnosis and the surrounding software stack. It spans 3,086 published records filed between 2015 and mid-2026, drawn from receiving offices led by the United States, India and the European Patent Office.
The scope sits at the intersection of mechanical prognostics and industrial software: a bearing degradation model is only patentable at scale once it is wired into a monitoring platform, a digital twin, or an industrial IoT pipeline, which is why control-systems and AI-model classifications dominate the composition figures below.
Filing trend and technology composition
Two views of the same 3,086-record dataset: how filing activity has moved year over year, and which IPC subclasses carry the claims.
Filings rose from 96 in 2017 to a peak of 375 in 2025
Volume climbed through the late 2010s and plateaued in the low-to-mid 2020s: 2021 (261) to 2024 (252) moved -3%, the last span long enough to read as complete given the roughly 18-month lag between filing and publication. 2026 figures (283 so far) are still filling in.
Control systems and AI models dominate the classification mix
G05B (control & regulating systems) appears on 40.3% of records and G06N (AI-based computing) on 35.3%, with G06Q business-process classes on 23.9% and G06F digital data processing on 20.5%. Because a record can carry several IPC codes, these shares add up to well over 100% of the 3,086 records in scope — the overlap itself is the signal: RUL prediction is claimed jointly with control logic and machine-learning method claims far more often than as a standalone sensing or measurement invention (G01M at 10.9%, G01R at 9.0%).
Shares are the percentage of the 3,086 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Predictive Maintenance — Bearing Remaining-Useful-Life Prediction Patent Landscape with Eureka
This page is one run against one query. Ask Eureka your own question about predictive maintenance — bearing remaining-useful-life prediction patent landscape and every answer comes back with the patent numbers behind it.
Try EurekaA representative filing and the most-cited prior art
Methods And Apparatus For Estimating Remaining Useful Life — Siemens Aktiengesellschaft (2026-03-12)
The filing describes training a neural network on a historical dataset of condition-monitoring readings paired with observed remaining-useful-life values for a worn part, with the network outputting a parameter of the RUL distribution rather than a single point estimate.Framing RUL as a learned probability distribution, not a point value, is the detail worth watching when assessing freedom to operate.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20150201918A1 | Surgical Handpiece | 1,097 |
| 2 | US20190339688A1 | Methods and systems for data collection, learning, and streaming of machine signals for analytics and mainten… | 974 |
| 3 | US20170286572A1 | Digital twin of twinned physical system | 767 |
| 4 | US20200348662A1 | Platform for facilitating development of intelligence in an industrial internet of things system | 733 |
| 5 | US20210157312A1 | Intelligent vibration digital twin systems and methods for industrial environments | 716 |
| 6 | US5311562A | Plant maintenance with predictive diagnostics | 658 |
| 7 | US20180284758A1 | Methods and systems for industrial internet of things data collection for equipment analysis in an upstream o… | 600 |
| 8 | US20200225655A1 | Methods, systems, kits and apparatuses for monitoring and managing industrial settings in an industrial inter… | 563 |
| 9 | US20200103894A1 | Methods and systems for data collection, learning, and streaming of machine signals for computerized maintena… | 514 |
| 10 | US20190033845A1 | Methods and systems for detection in an industrial internet of things data collection environment with freque… | 412 |
Citation counts favour older publications simply by virtue of longer exposure; treat them as a measure of influence within this corpus, not of present-day importance.
Each row carries its publication number; clicking a row searches Eureka by that number.
Put your own technology through the same analysis
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →What the data means for filing strategy
Three read-throughs from the concentration, composition and trend figures above.
One leader, then a long tail
The leading assignee alone holds 352 records against a fifth-place figure of 53 and a tenth-place figure of 36 — a steep drop-off after the top position. The top 5 combined reach 24.2% of all records in scope and the top 10 reach 30.6%, meaning roughly seven in ten records sit outside the ranked leaders entirely.
Controls-plus-AI is the default claim shape
G05B control systems and G06N AI-model classifications each cover well over a third of records, and they frequently co-occur on the same filing. A bearing RUL claim that leans purely on sensor physics without a control or learning layer is now the exception rather than the rule.
Occupied, not accelerating
Filing volume held roughly flat across the last fully-comparable span, 261 in 2021 to 252 in 2024. That reads as a mature, densely claimed space rather than a growth market — new entrants are more likely to find prior art blocking straightforward claims than open ground.
US and India lead receiving offices
The United States receives the largest share of filings (1,196), with India (638) and the EPO (359) next, followed by WIPO/PCT filings (277). A filing strategy built only around US and EPO coverage would miss the second-largest receiving office in this dataset.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to predictive maintenance — bearing remaining-useful-life prediction patent landscape, with the prior art for and against each one.
Where to take this next
The dataset points to a densely claimed core and a thinner set of adjacent branches. Two practical follow-ups worth running.
Map freedom-to-operate against the top portfolio
With one assignee holding 352 records versus a fifth-place figure of 53, any new RUL filing should first be checked against that single portfolio's claim scope before broader prior-art work begins.
Explore assignee claims in EurekaTest claims in the under-claimed branches
Federated and transfer-learning approaches to bearing RUL show up thinly relative to the dominant controls-plus-AI claim shape, suggesting narrower but more defensible filing opportunities.
Run a white-space search in EurekaCommon questions on this landscape
One assignee leads the field with 352 records, well ahead of the fifth-ranked filer at 53 and the tenth-ranked filer at 36. The top 5 assignees combined account for 24.2% of all 3,086 records in scope, and the top 10 combined reach 30.6%. That means the majority of filings sit outside this ranked leadership group, spread across a long tail of single- or few-filing entrants including established industrial-equipment makers and diversified technology firms.
Filing volume grew from 96 in 2017 to a peak of 375 in 2025, but the most recent fully comparable span (2021 to 2024) actually moved -3%, from 261 down to 252. Because publication lags filing by roughly 18 months, the 2025 and 2026 figures will revise upward over time, but the underlying signal is a plateau rather than continued acceleration. Treat this as a mature, densely occupied claim space rather than an emerging one.
The dominant classifications are G05B (control and regulating systems, 40.3% of the 3,086 records) and G06N (AI-based computing, 35.3%), often appearing together on the same filing. G06Q business-process classes cover 23.9% and G06F digital data processing 20.5%. Pure sensing and measurement classes such as G01M and G01R sit well below 11% each, showing that most claims wrap prediction logic into a control or software layer rather than staying at the sensor level.
The densest claim territory is controls-plus-AI framing of RUL prediction, so branches that sit outside that pattern show up thinner: edge-embedded prognostics that run the model on the bearing sensor itself, federated learning across a fleet's bearing data without centralising it, and physics-informed models that output a full RUL distribution rather than a point estimate. These are not guaranteed clear, but they are less crowded than the dominant claim shape and worth a targeted freedom-to-operate check before filing.
The United States receives the largest share of filings at 1,196, followed by India at 638 and the European Patent Office at 359. WIPO/PCT filings add 277, with Germany (89) and Australia (88) trailing well behind. A filing strategy focused only on the US and Europe would miss India, which is the second-largest receiving office in this dataset by a wide margin.
Research Predictive Maintenance — Bearing Remaining-Useful-Life Prediction Patent Landscape in depth with Eureka
Go past this page: query the whole predictive maintenance — bearing remaining-useful-life prediction patent landscape corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.
Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company’s registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.