Why do 95% confidence and 95% reliability need 59 parts?

Because 59 is the smallest count where 0.95 to that power drops below 0.05. A cavity making only 95% good parts would give 59 straight good parts less than 5% of the time.

Is 30 shots per cavity enough for a mold trial?

Not to prove 95% reliability from pass or fail results. Thirty good parts with zero rejects prove only about 90.5% at 95% confidence; 95% needs 59 per cavity and 99% needs 299.

Can I pool parts from all cavities to reach the sample?

Only if your claim is about the mold as a whole. A claim about each cavity needs the full sample from each, because a blocked vent can short-shoot one cavity while the others fill.

What if I find one reject during the trial?

Then the zero-failure claim fails at this sample size. Fix the cause and run a fresh sample, either this one or a plan that allows one reject: at 95% and 95% that takes 93 parts per cavity.

Injection Molding calculator

Mold Trial Sample Size Calculator

Size a mold trial or validation run before you book the press: parts to inspect per cavity so zero rejects prove your reliability target. You need cavities, confidence, reliability and any customer minimum.

What this calculator does

Parts to inspect from each cavity, and in total, so a zero-reject mold trial proves your reliability target.

Formula used

Zero-failure sample per cavity n = CEILING(ln(1 − confidence) ÷ ln(reliability))
Parts to inspect per cavity = MAX(n, minimum shots per cavity); parts to inspect = cavities × that
Shots to run = startup shots + parts to inspect per cavity
Parts molded in the trial = shots to run × cavities
Reliability the minimum alone proves = (1 − confidence)^(1 ÷ minimum shots per cavity)

Inputs explained

Cavities: Cavities filling every shot; each is sampled as its own process.
Confidence Level: From your validation plan or the customer's quality agreement.
Reliability to Prove: Share of good parts each cavity must be shown to make.
Minimum Shots per Cavity: Your customer's or quality plan's floor; enter 1 if none.
Startup Shots Before Sampling: Shots scrapped until cycle and cushion settle, from the setup sheet.

How to use the result

Best suited to writing a T1 or T2 trial protocol, sizing a validation run per cavity, checking a 30-shot habit against the plan.
Zero-failure plans only; a plan that tolerates rejects needs a larger sample. Proves a conforming rate, not capability; a Cpk study needs its own sample size.

Current U.S. benchmarks

The producer price index for plastic resins and materials stands at 280.569 (BLS, Aug 2026), up 6.6% from a year earlier. Quotes priced off last quarter's material cost miss this move.
The U.S. has 9,635 plastics product manufacturing establishments employing about 677,302 workers (Census County Business Patterns, 2023).

Common questions

Why do 95% confidence and 95% reliability need 59 parts? Because 59 is the smallest count where 0.95 to that power drops below 0.05. A cavity making only 95% good parts would give 59 straight good parts less than 5% of the time.
Is 30 shots per cavity enough for a mold trial? Not to prove 95% reliability from pass or fail results. Thirty good parts with zero rejects prove only about 90.5% at 95% confidence; 95% needs 59 per cavity and 99% needs 299.
Can I pool parts from all cavities to reach the sample? Only if your claim is about the mold as a whole. A claim about each cavity needs the full sample from each, because a blocked vent can short-shoot one cavity while the others fill.
What if I find one reject during the trial? Then the zero-failure claim fails at this sample size. Fix the cause and run a fresh sample, either this one or a plan that allows one reject: at 95% and 95% that takes 93 parts per cavity.

Last reviewed 2026-10-01.