Page 330 - Six Sigma Demystified

310 Six SigMa DemystifieD

Interpretation

The upper and lower control limits indicate the bounds of expected process
behavior. The fluctuation of the points between the control limits is due to the
variation that is intrinsic (built in) to the process. We say that this variation is
due to common causes that influence the process. Any points outside the con-
trol limits can be attributed to a special cause, implying a shift in the process.
When a process is influenced by only common causes, then it is stable and can
be predicted.
If there are any out-of-control points, then special causes of variation must
be identified and eliminated. Brainstorm and conduct designed experiments to
find the process elements that contribute to sporadic changes in process loca-
tion. To predict the capability of the process after special causes have been
eliminated, you should remove the out-of-control points from the analysis,
which will remove the statistical bias of the out-of-control points by dropping
them from the calculations of the average and control limits.
Note that some SPC software will allow varying sample sizes for the Np chart.
In this case, the control limits and the average line will be adjusted for each
sample. Frequently, it is less confusing to use a P chart for these data because only
its control limits will vary (the average line will remain constant). See “Run-Test
Rules” and “Statistical Process Control (SPC) Charts” for more detail.

P Chart

A P chart is one of a set of control charts specifically designed for attributes data.
The P chart monitors the percent of samples having the condition, relative to

either a fixed or varying sample size, when each sample either does have this
condition or does not have this condition. For example, we might choose to look
at all the transactions in the month (since this would vary from month to month)
or a set number of samples, whichever we prefer. From this sample, we would
count the number of transactions that had one or more errors. We then would
track on the P control chart the percent of transactions with errors each month.

When to Use

Measure Stage
• To estimate, using attributes data, the process baseline (Generally, we
would greatly prefer to use variables control charts for this purpose.)