The ranked list in this dataset covers 100 companies, and it is concentrated rather than evenly spread: the top 5 combined hold 456 of the 1,442 records in scope, or 31.6%. The single leading assignee holds 308 records, more than ten times the count at tenth place, which is a wide gap for a mechanical engineering field. Below the top five, filing counts drop off quickly into a long tail of assignees with only a handful of records each, so most of the ranking should be read as opportunistic filers rather than sustained portfolio builders.

Filings rose from 37 in 2021 to 57 in 2024, a 54% increase over that three-year span, which is the clearest recent trend the data supports. Counts appear to taper after 2024, but that reflects the roughly 18-month lag between when a patent is filed and when it publishes, not a genuine drop in activity. Treat 2024 as the most recent complete year and be cautious drawing conclusions from 2025 or 2026 figures until more publications catch up.

US20190316989A1, filed by FutureMain Co., Ltd, describes a method for predicting faults in rotary machine elements by coding fault conditions into four categories — fabrication and installation, load, lubrication, and environment and operation — and matching incoming status data against a stored association database. It is representative of a broader pattern in this dataset where fault-mode coding and prediction, rather than the mechanism’s physical structure, is the subject of the claims. Anyone building a fault-prediction scheme for rotary equipment should check their specific coding taxonomy and data-matching approach against this filing and its family rather than assume mechanical novelty alone clears it.