How do you calculate process validation sample size?

Multiply the number of validation runs by the sampling rate per run, then take the larger of that figure and your minimum samples per run. With 3 runs at a 10% rate but a 30-sample minimum, the minimum governs and the required sample size is 30 samples.

What is a good sample size for process validation?

For many Class II devices, validation teams target 30 samples per run as a practical floor, or sizes that support a 95% confidence / 95% reliability claim for critical attributes. The right number scales with risk classification and the defect rate you must rule out.

Why use a minimum samples per run at all?

A rate alone can collapse to a handful of units on short runs, which is indefensible to an auditor. The minimum floor (30 here) ensures every validation run carries enough statistical weight regardless of batch size.

How many PPQ runs do I need?

Three consecutive successful runs (the basis for the default of 3) remains the conventional expectation for traditional process validation, though a lifecycle/continuous approach may justify a different count with documented rationale.

Is this the same as a 95/95 reliability sample size?

No. A 95/95 plan is derived from binomial or normal tolerance statistics for a specific confidence and reliability. This calculator sizes by rate and floor, so confirm critical-to-quality attributes against a formal statistical table.

Does a higher sampling rate always increase the sample?

Only once the rate-driven count exceeds the minimum. At 3 runs and 10%, the rate result sits below 30, so the 30-sample floor controls until you raise runs or rate enough to clear it.

MedTech Manufacturing calculator

Process Validation Sample Size Calculator

Process Validation Sample Size tells a medical device manufacturer how many units must be inspected across IQ/OQ/PQ runs to demonstrate a process is in a validated state of control. Quality and validation engineers use it when building PPQ protocols under FDA 21 CFR 820.75 and ISO 13485 to justify their sampling plan to auditors. It matters because too few samples leave you unable to defend a confidence/reliability claim, while a minimum floor guarantees you never run a validation on a statistically trivial sample. The result is the larger of your rate-driven sample count and a hard minimum, so the floor protects low-volume runs.

What this calculator does

Estimate total sample count for process validation from number of PQ runs, sampling rate per run, and minimum per-run requirement.
Use this when writing validation protocols, planning lab capacity for PQ activities, or estimating material consumption for validation batches.
It multiplies the number of validation runs by the per-run sampling rate, then returns the larger of that result and your minimum samples per run.

Formula used

Calculated validation samples = number of validation runs × sampling rate per run
Required process validation sample size = max(calculated samples, minimum samples per run)

Inputs explained

Number of validation runs: Consecutive PQ runs required (typically 3 per FDA/ISO guidance; may be more for high-risk devices).
Sampling rate per run: Percentage of batch sampled for testing at each validation run (based on statistical requirements).
Minimum samples per run: Minimum sample count per run for statistical validity (often 30+ for attribute data per ANSI Z1.4).

How to use the result

Use it while drafting a PPQ or OQ/PQ sampling plan to size the inspection sample before locking the validation protocol.
It does not derive a statistically rigorous confidence/reliability sample size (e.g., 95/95 from a binomial or normal tolerance table) — it sizes against a rate plus a floor, so for high-risk attributes confirm against an ANSI/ASQ Z1.4 or C=0 plan.

Current U.S. benchmarks

U.S. manufacturing runs at 75.6% of capacity with new factory orders at $657B per month (Federal Reserve and Census, May 2026).
The U.S. has 8,825 medical equipment and supplies establishments employing about 308,388 workers (Census County Business Patterns, 2023).

Common questions

How do you calculate process validation sample size? Multiply the number of validation runs by the sampling rate per run, then take the larger of that figure and your minimum samples per run. With 3 runs at a 10% rate but a 30-sample minimum, the minimum governs and the required sample size is 30 samples.
What is a good sample size for process validation? For many Class II devices, validation teams target 30 samples per run as a practical floor, or sizes that support a 95% confidence / 95% reliability claim for critical attributes. The right number scales with risk classification and the defect rate you must rule out.
Why use a minimum samples per run at all? A rate alone can collapse to a handful of units on short runs, which is indefensible to an auditor. The minimum floor (30 here) ensures every validation run carries enough statistical weight regardless of batch size.
How many PPQ runs do I need? Three consecutive successful runs (the basis for the default of 3) remains the conventional expectation for traditional process validation, though a lifecycle/continuous approach may justify a different count with documented rationale.
Is this the same as a 95/95 reliability sample size? No. A 95/95 plan is derived from binomial or normal tolerance statistics for a specific confidence and reliability. This calculator sizes by rate and floor, so confirm critical-to-quality attributes against a formal statistical table.

Last reviewed 2026-05-12.