The Reliability and Confidence Sample Size Calculator
The Reliability and Confidence Sample Size Calculator will provide you with a sample size for design verification testing based on one expected life of a product. This calculator works by selecting a reliability target value and a confidence value an engineer wishes to obtain in the reliability calculation. Quality-One uses this calculator to intelligently manage the performance risk of a new product or process design in the design verification or validation process.
Using the Calculator for Design Verification
Design Verification Plan and Report (DVP&R) requires a sufficient sample size to justify performance inferences about a design. Sample sizes of 1 are typically used due to the high cost of prototypes and long lead times for testing. With a sample size of 1, it will be very difficult to determine where the distribution is located or the type of distribution indicated. There is always risk involved when selecting a sample size for testing. Recognizing the reliability and confidence is a key step in mitigating the performance risk in Design Verification and Validation.
The Binomial Distribution
The calculator is based on discreet distribution known as the Binomial Distribution.
P(x)= nCxpx(1-p)n-x
Where:
n = sample size
x = item of interest
p = probability or proportion defective
The Binomial Distribution is used to determine acceptance of a product in a defined set of discreet circumstances:
- All tests must be identical
- Each test has two possibilities – Success or Failure
- Probability of pass or fail for each test does not change from test to test
- The outcome of one test does not affect the outcome of any other test
We can apply the Binomial Distribution in Design Verification because each of the prerequisites listed above must also be true when testing prototypes to a pass / fail criteria.