 |
Carillon Technologies Limited |
Zero Defect Sampling |
|
|
This Zero Defect Sampling page is based on the use of the Poisson Distribution that is based on the premise that we are sampling from a large lot size and the sample size is small in comparsion.
The Poisson Distribution is defined with parameters
r = acceptance number
n = sample size and
p = fraction nonconforming
P(r) = Probability of accepting a lot with r nonconforming
| The Poisson distribution mathematically can be represented as: |
 |
| For Zero Defect Sampling we substitute r = 0 into the Poisson expression |
 |
The Excel Worksheet (probaccept.xls) below below provides us with examples of the effectiveness of our zero defect sampling plan for various n's and p's.
|
|
||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| If we plot the data from the previous table for sample size of 20 then we have the representation of the effectiveness of the sampling plan better known as the OC Curve. |
 |
By solving for n or p we can provide ourselves with a useful predictive formulas.
 |
 |
|
Using the formula for p we can solve for the appropriate sample size using the P(a) of 0.10 or 10% (In Mil Std 105E this is called the Limiting Quality) This Excel Worksheet calculate the necessary sample size to reject p with 90% certainity. Link to Excel Worksheet Calculator (samplesize.xls) (you must have Excel open on your computer and your browser preferences set to open excel when .xls is linked) |
|
