First article inspection is the most documentation-intensive measurement task in a Tier-1 supplier's quality workflow. For a typical stamped and machined part with 40–80 drawing characteristics — flatness, profile, true position, surface finish, hole diameters — the FAI generates dimensional measurement data for every characteristic on five to ten representative parts, plus a material certificate, a cosmetic inspection record, and the completed PPAP submission documents. When that dimensional data comes from a CMM queue with a two-week backlog, PPAP timelines stretch accordingly.
Automated vision measurement cannot replace a CMM for every FAI characteristic — and we are not arguing that it should. But for the subset of features where a calibrated vision system's measurement uncertainty is within the AIAG MSA acceptability threshold for the given tolerance, vision-measured data is fully valid for inclusion in a dimensional results report. The question worth examining is which characteristics that actually covers, and how to document the capability claim correctly.
What the AIAG PPAP Manual Requires for Dimensional Results
The AIAG PPAP manual (4th edition) requires a dimensional results report (DRR) showing measurements of all drawing characteristics on the number of parts specified by the customer (typically five parts minimum for Level 3 PPAP). Each characteristic requires: the engineering specification, the actual measurement value, and a pass/fail disposition. The DRR must be signed by the responsible quality engineer and traceable to the measuring equipment used — including equipment ID and calibration status.
Critically, the PPAP manual does not prescribe what measurement method must be used. It requires that the measurement system be included in the supplier's measurement system analysis program (per the AIAG MSA reference manual) and that the measurement uncertainty of the system is documented and accepted as adequate for the tolerance being measured. A vision inspection system that passes a Gauge R&R study against CMM reference measurements meets this requirement for the characteristics studied.
This is the technical basis for including vision measurement data in a PPAP DRR: not a claim that vision is equivalent to CMM, but a documented capability study demonstrating that vision measurement uncertainty is acceptable for the specific features and tolerances in question.
FAI Feature Categories: Where Vision Contributes
Not all FAI characteristics are equal from a measurement method standpoint. A practical framework for categorizing features by vision suitability:
High vision suitability — features where 2D or 3D vision measurement uncertainty is routinely below 10% of the tolerance (the AIAG MSA guideline for attribute gauge acceptability used as a starting reference):
- Hole diameter and location (2D area-scan, sub-pixel edge detection) for tolerances above
±0.1mm - Profile of a surface (structured-light 3D) for tolerances above
0.3mm - Flatness (structured-light 3D) for tolerances above
0.2mm - Linear dimensions and edge-to-edge distances (area-scan, calibrated field) for tolerances above
±0.15mm - Surface defect classification (area-scan AOI) per control plan defect classes
Conditional vision suitability — features where capability depends on part-specific geometry and surface condition:
- True position of features relative to datum scheme — vision can measure position but datum establishment from datum features requires careful fixture design to match CMM datum contact
- Parallelism and perpendicularity — achievable with structured-light 3D but sensitive to part fixturing consistency
- Angularity — measurable on flat features, less reliable on complex freeform surfaces
CMM-required features — where vision measurement uncertainty is too large for the tolerance, or the feature is physically inaccessible to optical measurement:
- Internal bore diameters (thread gauging, deep blind holes)
- Sub-
0.05mmtolerances on precision machined surfaces - Runout on rotating features requiring rotary datum fixturing
The CMM Queue Problem and Parallel Measurement Strategy
Consider a scenario that is representative of regional Tier-1 plants in Southeast Michigan: a new door hinge bracket program requires PPAP Level 3 submission. The part has 62 drawing characteristics. The CMM queue at the plant is running at an average 10-business-day backlog due to production measurement demands. The program timing requires PPAP submission within three weeks of first article run.
With a 10-day CMM queue and three-week window, the timeline is technically achievable — but leaves no buffer. If the first-run parts have any out-of-specification characteristics requiring second-run iteration, the program misses.
The parallel measurement approach: run vision measurement on the 42 high-suitability features immediately after the first article run. Vision measurement of 42 characteristics on five parts takes roughly 2–4 hours in a calibrated inspection cell, compared to the 10-day CMM queue wait. The CMM queue is then reserved for the 20 features in the conditional and CMM-required categories.
This compresses the effective measurement timeline for two-thirds of the characteristics from 10 days to 4 hours, while maintaining CMM measurement for the features that require it. The DRR reflects which characteristics were measured by vision (with equipment ID and Gauge R&R reference) and which by CMM.
Gauge R&R Documentation for the FAI Package
Including vision measurement data in a PPAP DRR requires a Gauge R&R study on file that covers the vision system, the specific features measured, and the tolerances in question. The study must follow the AIAG MSA reference manual methodology — typically the crossed Gauge R&R study with at least two appraisers (or, for automated systems, a repeatability-only study documented as a Type I gauge study).
For an automated vision inspection system without operator variation in the measurement act (the system measures the same way regardless of who triggers it), the relevant MSA metric is repeatability: the variation in the measurement result when the same part is measured multiple times under identical conditions. The standard AIAG MSA threshold is %GRR < 10% of the tolerance for acceptance, with 10–30% requiring engineering judgment based on feature criticality.
The Gauge R&R study should be part of the vision system's IQ/OQ/PQ validation documentation — not generated fresh for each PPAP. When the vision system is already qualified under IQ/OQ/PQ covering the feature types and tolerance ranges relevant to a new program, the existing MSA documentation can be referenced in the PPAP package with a supplemental note confirming that the new part's features and tolerances fall within the qualified measurement range.
Generating the DRR Output Format
The AIAG PPAP DRR has a defined structure: part name and number, revision level, supplier PPAP number, characteristic number per the drawing balloon, engineering specification, measurement result, and pass/fail. Most quality engineers maintain DRR templates in Excel or their QMS (SAP QM, Oracle Quality, or IATF-aligned QMS platforms).
Vision inspection systems that output measurement results in structured CSV or XML format can populate the DRR template directly with a mapping script or QMS integration, eliminating manual transcription. The equipment ID and calibration certificate number are appended automatically from the system configuration record.
What auditors and customer SQEs look for in a vision-measured DRR is the same as for any other measurement method: traceability of the measurement result to the measuring instrument, and evidence that the measurement instrument is qualified for the feature and tolerance. The measurement method itself (vision versus CMM versus manual gauge) is secondary to the documentation of capability.
What FAI Automation Does Not Replace
We are not saying that vision-based FAI measurement eliminates the need for CMM measurement or for the quality engineer's judgment in the PPAP process. For new programs with complex GD&T datum schemes, the CMM measurement of datum features and the critical-path characteristics is still required — and the FAI process as a whole requires the quality engineer to review all measurement data, sign off on the DRR, and submit the PPAP package.
What vision measurement contributes is a faster first read on the large category of measurable features, giving the quality engineer more time to focus on the complex characteristics rather than waiting in a CMM queue for data on dimensions that a calibrated vision cell could have reported the same day as the first article run.
For suppliers running two or more new PPAP submissions concurrently — which is common for growing regional Tier-1 plants managing multiple customer programs — the cumulative time savings across the DRR process directly affect program timing risk.
Qcvisionly's pilot scope includes the Gauge R&R documentation for vision-measured FAI features and the DRR output integration with your QMS. If you are managing upcoming PPAP submissions with CMM queue constraints, request a pilot discussion to review which features in your programs fall within vision measurement capability.