Research study · precision manufacturing

When a part fails FAI, which balloon does the evidence belong to?

Five founder-authored synthetic cards follow one made-up inspection workflow from ballooned drawing characteristics to measurement evidence to a FAI / AS9102 package.

Answer one 60-second question About this research
These are founder-authored synthetic workflow scenarios. They are not customer data, not product UI, not product proof, and not evidence of demand. No confidential, ITAR, CUI, export-controlled, customer-restricted, or proprietary documents are needed.
SYNTHETIC PART · DWG ST-001 · REV B STATIC ILLUSTRATION

All five balloons and their evidence status are shown together below.

88.90 +/-0.10 OAL 1 2 3 4 5
BLN 1FEAT_BORE_B01_DIA = 12.71 · PASSCLEAR MATCH
BLN 2FEAT_LEN_OAL_REVA = 88.86 · in tolerance, Rev AREV MISMATCH
BLN 3FEAT_FLAT_A_DAT = 0.031 vs 0.02 · FAILOUT OF TOL
BLN 4HOLE_A12_POS / HOLE_A13_POS · both plausibleAMBIGUOUS
BLN 5Ra 0.8 requirement · no linked evidenceMISSING
PACKAGE STATUS: 5 characteristics · 1 ready · 4 holds.
AS9102 Form 3FAIRGD&TRevision controlCMM rowNCRRa 0.8Balloon to evidence

What you are looking at

A synthetic scenario, built to be corrected

The part, drawing numbers, tolerances, CMM rows, and statuses are invented. The point is not to demonstrate a tool; it is to make the inspection-documentation question concrete enough for a practitioner to correct.

These are founder-authored synthetic workflow scenarios. They are not customer data, not product UI, not product proof, and not evidence of demand.

The workflow gap

Three documents describe the same feature. They still have to be reconciled.

Each dimension can appear as a requirement on a drawing, a row in a measurement report, and a line in a FAI package. These synthetic cards ask where that link breaks or slows down.

1

The drawing requirement

A ballooned characteristic with a tolerance, datum context, and revision state.

BLN 3 · FLATNESS 0.02 | A REV B · released drawing
2

The measurement evidence

A CMM report, gauge log, or supplier certificate named by feature, instrument, or program convention.

CMM ROW "FEAT_FLAT_A_DAT" = 0.031 status FAIL · mapped to BLN 3
3

The package consequence

Form 3 needs defensible evidence for each characteristic before the package can move forward.

AS9102 FORM 3 · CHAR 3 HOLD — review required before completion

The five cards

From ballooned characteristic to package consequence

The five cards cover a deliberately broad failure spectrum: clear match, revision mismatch, out of tolerance, ambiguous match, and missing evidence. Select a card to expand it.

One named CMM row clearly maps to balloon 1 and is in tolerance. This is the clean baseline.

DrawingBore diameter 12.70 +/-0.05.
EvidenceFEAT_BORE_B01_DIA = 12.71 PASS.
PackageResult can be entered without reconciliation.
Synthetic scenario card: a ballooned drawing excerpt for a bore diameter of 12.70 plus or minus 0.05, one CMM row FEAT_BORE_B01_DIA measuring 12.71 with status PASS, and a package panel showing the result is ready to enter the FAI / AS9102 package.
Synthetic illustration — invented part and data, not customer material.

Founder-operated demo

The browser walkthrough is not public yet

A short interactive walkthrough of these five synthetic cases exists as a founder-operated demo. Ask by email if you want to see it.

What we are validating

Two directions of the same link. Which one creates more avoidable work?

Both problems live between the balloon and the evidence. The research question is deliberately open so respondents can reject this framing or point to something else.

A · FAI prep

Preparing the FAI

  • Ballooning the drawing and extracting characteristics into Form 3.
  • Planning which instrument or program covers each balloon.
  • Catching missing planned evidence before package assembly.
PLAN COVERAGE · SYNTHETIC BLN 1-4 CMM PROGRAM planned BLN 5 SURFACE FINISH Ra 0.8 no instrument assigned
B · Matching back

Matching results back to balloons

  • Mapping CMM rows, gauge logs, and certificates back to balloon numbers.
  • Flagging evidence captured against a superseded revision.
  • Resolving ambiguity when multiple rows fit the same balloon.
EVIDENCE MAP · SYNTHETIC FEAT_BORE_B01_DIA → BLN 1 clear match FEAT_LEN_OAL_REVA → BLN 2 Rev A capture FEAT_HOLE_A12/13 → BLN 4 ambiguous Ra 0.8 → BLN 5 no candidate

Who we want to hear from

People who sign the forms, run the measurements, or fix the package.

Quality managers

You own the FAIR package and know where review time concentrates.

Metrology / CMM

You know how feature names, programs, and result tables get interpreted later.

Manufacturing engineers

You see the impact when revision or evidence gaps surface late.

Supplier quality

You reconcile supplied evidence against drawing and customer package expectations.

The one question

Does this match how the work actually goes?

A one-line email answer is useful, especially if the honest answer is that these cards emphasize the wrong thing. There is nothing to buy and nothing to sign up for; this is a research conversation.

Answer one 60-second question