Precision - Accuracy
Is the accuracy of the measurement you plan
to make. This is relevant to the scale and/or detail of your operational
definition or description but could also have an influence on the sample
size being measured. In any six sigma
or ISO-9000 plan, adequate precision
is required to monitor process
improvements and for the baseline
measurements in any benchmarking
objectives.
For an example we will set up a test with a
histogram plot control
chart using the following sample.
| Part
Number |
Revision |
Part Description |
| Precision |
A |
Accuracy Test |
Here is the parameter
information.
| Parameter |
Print
Loc. |
Min.
Spec. |
Max.
Spec. |
Nom.
Spec. |
| Accuracy |
A |
1 |
10 |
5.5 |
Using only one decimal point, our chart
looks like this.
| n |
x1 |
x2 |
x3 |
x4 |
x5 |
X-Bar |
Range |
Cp |
Cpk |
| 1 |
5.5 |
5.5 |
5.5 |
5.4 |
5.6 |
5.5 |
0.2 |
21.213 |
21.213 |
| 2 |
5.4 |
5.4 |
5.4 |
5.5 |
5.6 |
5.4 |
0.2 |
16.77 |
16.621 |
| 3 |
5.6 |
5.6 |
5.4 |
5.5 |
5.6 |
5.5 |
0.2 |
16.77 |
16.621 |
| 4 |
5.5 |
5.6 |
5.4 |
5.4 |
5.6 |
5.5 |
0.2 |
15 |
15 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Totals |
21.9 |
0.8 |
69.753 |
69.455 |
|
|
|
|
|
Avg. |
5.5 |
0.2 |
17.438 |
17.364 |
Using only two decimal points, our chart
looks like this. All we did was modify the original data by a couple of
tenths.
| n |
x1 |
x2 |
x3 |
x4 |
x5 |
X-Bar |
Range |
Cp |
Cpk |
| 1 |
5.54 |
5.56 |
5.55 |
5.38 |
5.58 |
5.52 |
0.2 |
18.576 |
18.485 |
| 2 |
5.41 |
5.41 |
5.39 |
5.4 |
5.65 |
5.45 |
0.26 |
13.514 |
13.369 |
| 3 |
5.58 |
5.59 |
5.48 |
5.55 |
5.61 |
5.56 |
0.13 |
29.588 |
29.18 |
| 4 |
5.54 |
5.65 |
5.45 |
5.45 |
5.65 |
5.54 |
0.2 |
14.985 |
14.825 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Totals |
22.07 |
0.79 |
76.663 |
75.859 |
|
|
|
|
|
Avg. |
5.52 |
0.2 |
19.166 |
18.965 |
The additional accuracy extends to control
charts to include X-Bar and R-Bar.
Also see Quality
inspection, DMAIC,
(Define, Measure, Analyze, Improve), Control, R
& R, (Repeatability and Reproducibility), Tolerance
Design, and Variable
data.
|
