Lattice24 · Technical brief
A failed sensor alarms. A drifted sensor does not.
It reports a plausible wrong number, and the control room has no way to know. That is why calibration runs on a schedule rather than on demand — nobody can tell which instrument has moved until someone brings a reference to it. Between calibrations you are running on numbers whose error is unknown and unbounded.
We read your existing sensor stream through a second, independent estimator and watch the two readings separate. The disagreement tracks the sensor's true error — with no reference instrument involved.
Worked example · ethanol · month 3
The size of the failure was called correctly by a system that never saw the reference.
| Product | r | 95% CI | Holds |
|---|---|---|---|
| Ethylene | +0.983 | +0.806 → +0.997 | Yes |
| Ethanol | +0.914 | +0.144 → +0.992 | Yes |
| Acetone | +0.789 | +0.425 → +0.947 | Yes |
| Toluene | +0.581 | −0.850 → +1.000 | Too few points |
| Ammonia | +0.100 | −0.684 → +0.902 | No |
Three of five products hold, with intervals entirely above zero. Ammonia does not — an earlier five-batch estimate of +0.562 did not survive testing across all ten batches and is withdrawn here rather than quietly dropped.
The second estimator does not need to be more accurate than your sensor. It needs to fail differently. Two independent readings of the same process drift apart in a way a single reading cannot reveal, and the size of the gap carries the information.
That is why the method needs no calibration gas, no reference cell and no downtime — only the data you are already logging.
Stated plainly, because these were tested and did not hold.
The underlying engine has been run across 100 measurement domains and a blind random draw of open scientific datasets, with all results — including failures — published under CC-BY-4.0. Oil & gas domains from that run, one unmodified pipeline, no per-domain tuning:
| Domain | Channels | Signal | p |
|---|---|---|---|
| Compressor station | 25 | 23/25 | 5.2e-06 |
| Gas distribution | 24 | 24/24 | 1.2e-07 |
| Drilling rig | 15 | 14/15 | 8.4e-03 |
| Refinery unit | 8 | 8/8 | 7.8e-03 |
| Pipeline pressure | 9 | 8/9 | 1.3e-01 |
| Gas turbine | 11 | 9/11 | 8.3e-02 |
| Wellhead | 10 | 1/10 | 3.9e-03 |
| Pump bearing | 10 | 0/10 | 2.0e-03 |
Wellhead and pump bearing invert — their readings are far more regular than a randomised surrogate. That is the signature of a tightly controlled or steady-load system, and the direction of the inversion distinguishes a controlled process from an uncontrolled one. Both are reported here rather than omitted.
If the dates line up, the method works on your plant. If they do not, you have lost nothing but a data export — and we will say so.