Most filings in this space claim a method or system for isolating a section of a water distribution network — the district metered area, or DMA — and using flow measurements at its boundary to infer leakage or abnormal consumption. The claims typically cover how inflow is balanced against modelled or metered demand, and how boundary valves or meters are configured to make that isolation hydraulically reliable. In this dataset, 31.0% of the 29 records also carry an E03B water-supply-and-distribution classification alongside the measurement classes, reflecting that the physical network configuration is often claimed together with the analysis method.

Minimum night flow analysis specifically uses the low-demand overnight window, when legitimate consumption is at its lowest, to isolate background leakage and small bursts from the flow signal entering a DMA. General leak detection can rely on pressure transients, acoustic sensing or vibration monitoring at any time of day. The search behind this landscape combines minimum night flow with flow balance, burst detection and background leakage terms specifically, which is why testing/structure-balance (G01M) and water-supply (E03B) classes dominate over purely acoustic or vibrational method classes.

The ranking returned for this search covers 13 assignees, with the top-ranked filer holding 7 of the ranked filings and a sharp drop to 2 filings by fifth place and 1 by tenth. That shape — a single leader over a long tail — suggests the field is not yet dominated by a small group of blocking portfolios, though the leader's position is significant enough to warrant a direct claim comparison before filing something similar. The mix includes academic groups, specialist leak-analytics vendors and at least one large diversified technology company.

Filing peaked at 9 in 2019 and, across the 2021-to-2024 window this dataset can treat as filing-complete, moved from 1 filing to 0, a -100% change. That looks like a genuine lull rather than the field closing outright, and figures for 2025 and 2026 are still incomplete because patent publication typically lags the actual filing date by around 18 months. A shorter apparent decline in the most recent one or two years of any patent trend should always be read with that lag in mind rather than as a firm signal of falling interest.

Based on the class composition in this dataset, boundary-valve state verification, meter-accuracy drift correction and AI-based background-leakage estimation each appear across only a handful of the smaller subclasses (G01D, G01R, G06N each cover 6.9% to 17.2% of the 29 records) rather than forming a dense cluster. That suggests these are narrower, less-claimed branches compared with the bundled flow-balance-plus-analytics claims that dominate G01M and G06Q. A first claim drawn tightly around one of these narrower mechanisms, rather than repeating the common bundled pattern, is more likely to clear existing art cleanly.