Machine Vision & Industrial Inspection AI worked example

AI Inspection Sample Size at 99% vision system availability during validation: a worked example

What does the result look like when vision system availability during validation reaches 99%? The full calculation is worked below with real intermediate numbers. Use it when planning an AI inspection model validation study and you need to know how many parts to schedule and what proportion of those parts should be conforming versus seeded defectives.

The inputs for this scenario

  • Number of validation runs planned: 4 runs (unchanged)
  • Parts inspected per validation run: 250 parts / run (unchanged)
  • Vision system availability during validation: 99 % (raised for this scenario; the documented default is 88)
  • Proportion of conforming parts in the sample: 80 % (unchanged)

Working through the calculation

  • Applying the documented formula (Total planned parts = validation runs x parts per run) to the inputs above produces each figure below.
  • At this operating point the engine returns 792 parts for conforming parts in validation sample, the number this scenario is built around.
  • At this operating point the engine returns 1,000 parts for total planned validation parts.
  • At this operating point the engine returns 10 parts for parts lost to system unavailability.
  • At this operating point the engine returns 198 parts for defective parts in sample.

How this compares with the baseline

  • Against the tool's baseline example, where vision system availability during validation sits at 88% and the headline result is 704 parts, this scenario comes in 12.5% above the baseline at 792 parts.
  • A figure at this level is achievable when vision system availability during validation is genuinely sustained, not just peaked for a shift. It assumes a constant conforming proportion and a flat availability factor; real runs cluster downtime and defects, so treat the splits as planning estimates.

Results at a glance

  • Conforming parts in validation sample: 792 parts (headline result)
  • Total planned validation parts: 1,000 parts
  • Parts lost to system unavailability: 10 parts
  • Defective parts in sample: 198 parts

Run it with your numbers

  • Every input above is editable in the live AI Inspection Sample Size calculator, which recalculates instantly and can be shared with the inputs intact.

Last reviewed 2026-05-12.