In your process
Built into your
reliability loop.
You already have the loop. FMEA into DVP&R, DVP&R into execution, execution into RCA, RCA into corrective action, and back. Sidekick and Weave do not replace any of it. They close the gaps where it leaks.
- FMEA/
- DVP&R/
- Execution/
- RCA/
- 8D / CAPA
FMEA / FMECA
S × O × D = RPN
Sidekick is a detection control.
Detection is the one term in the RPN you can improve without redesigning the part or changing how often it fails. A mode you cannot see gets a poor detection rating and stays high on the list. Continuous watch on the value that reveals it lowers that rating honestly, so the modes you ranked highest are the ones that actually get caught.
Where it leaks today: The mode you rated dangerous is the one nobody was watching at 3am.
DVP&R
Verify each mode
Results log themselves, traceable to the mode they verify.
Every reading is captured off the instrument and attached to the run that produced it, so the line in the matrix and the evidence behind it are the same record. A proven setup becomes a template the next rig reuses, which is what makes the second programme cheaper than the first.
Where it leaks today: The matrix says pass. The evidence is in a spreadsheet somebody kept locally.
Execution
EVT · DVT · PVT · HALT · soak · HIL
The mode is caught the instant it triggers.
Live fault and state detection, out of band, so it does not depend on anyone being at the bench or on the rig's own machine still being alive. A soak that aborts at hour 3 does not quietly burn until morning.
Where it leaks today: Forty hours of chamber time spent on a run that died in the third.
RCA
5-whys · fishbone · FTA
The evidence package is already assembled.
An Observation carries the screen, the telemetry window and the exact config on the unit at the moment of failure. The analysis starts from the full record rather than a blank page, and the conversation never reaches can you reproduce it.
Where it leaks today: The root cause is unfindable because nobody recorded what the unit was running.
8D / CAPAsoon
Corrective action
What changed is on the record, and the fix is provable.
The config ledger shows what was on the unit and when it changed. Golden-run comparison shows the behaviour before and after. That is a corrective action an auditor can follow, rather than an assertion that it is better now.
Where it leaks today: The fix worked, and six months later nobody can show why.
And then it feeds back.
Observed occurrence and detection data flow back into the FMEA. The ratings stop being a workshop opinion from the start of the programme and start being what the floor actually saw. Reliability grows, and it compounds.
Who this page is for.
If you own the DVP&R, chair the FMEA reviews, or sign the 8D, this is the part of the product that matters to you. The box on the rig is the mechanism. This is the consequence.
Start where the loop leaks worst.
Pick the rig that fails silently or ties up an engineer. We put a Sidekick on it, read-only, and you see the floor view in weeks.