Process Evaluation
Worksheet Features
Evaluating whether a process is in or out of control is essentially a matter of deciding on whether
observed control points are in or out of control. The occurrence of an out of control point is called
an
out of control event. Since control limits are set up in accordance with some desired confidence level or equivalent, the rejection of an event as falling outside the process control limits is made with a definable level of rejection confidence.
The Process Evaluation worksheet is used to review the confidence for rejecting a process as being out-of-control
at a
selected point. The bottom half of the worksheet contains a table that
lists the control point data, the deviation from mean or reference
standard, the measurement process uncertainty, rejection confidence and
beta risk. Control points are also listed by date on the upper right
portion of the worksheet. Selecting a control point from this list
will display additional information on the upper left portion of the
worksheet.
Beta Risk Analysis
The Process Evaluation worksheet shows the probability that the point lies within the control limits, given the estimated
uncertainty in the measurement process. This in-tolerance probability is called the
beta risk. In cases where the uncertainty in the measurement process is appreciable, relative to the control limits, the beta risk can be critical in deciding whether to reject a process as being out-of-control. In short, even if a point lies outside the control limits, if the beta risk is high, the process should not necessarily be rejected.
Outlier
Identification
On occasion, it may be desired to exclude certain control points from analysis without deleting them from process history.
This occurs, for example, when one or more control points are suspected as being non-representative of the process. One way to determine whether a suspected point is non-representative is to see what impact it has on the process control chart's linear curve fit and projected intercepts. This impact can be assessed by excluding the suspected point.
One or more control points can be excluded by clicking the Exclude button adjacent to the control point in question.
If the Process Control Option is Parameter Interval Analysis, clicking this button will activate the Resubmission
Time Analysis Worksheet. Otherwise, clicking the button will cause it to display a check
mark.
Excluded
points become inactive in the process control history and are displayed in
gray on the Process Control Chart. These points are not included in developing process curve fits or in interval analysis. Likewise, they
do not appear in viewed or printed reports. All excluded points may be recovered by clicking Undo Excluded on the Reject Confidence menu.
Reject Confidence Report
A report showing rejection confidence, beta risk and other pertinent information for each control point can be generated by selecting the Print option from the
Process Evaluation Worksheet Print menu.
This report also indicates whether control points are in- or out-of-control, relative to the process control limits.
|