PRE-Control: 2-sided specification, nominal is best.

Note: in the downloadable spreadsheet, each cell will turn green, yellow, or red as the measurement is entered. This provides the operator or inspector with an immediate visual control. As an example, two measurements in the yellow zone will immediately show up as yellow. This feature (conditional cell color) does not appear to work in the Web version, although the Results column shows whether the sample (or qualification run) passes or fails.

The purpose of the upper and lower data limits is as follows. The spreadsheet cannot apparently tell a blank from a zero, so it would interpret a zero as well below the lower specification limit (for the nominal-is-best case) and flag the cell as a failure. To prevent this, the test condition for a Red (out of specification) result is anything between the Yellow regions and the data limits. Set the data limits so it is inconceivable that a legitimate measurement will never be outside them. The result is an additional visual control (error proofing) because a mistake, such as 9 instead of 90, will result in a visible entry in a non-colored cell (in the downloadable spreadsheet). All legitimate measurements will turn their cells either green, yellow, or red!

To try this out on the interactive spreadsheet below, enter your own values for the specification limits and the data limits. Then erase the sample entries and begin to enter your own. You will see how the spreadsheet counts the Green, Yellow, and Red measurements, and determines whether the sample passes or fails.

Here is a screenshot of what the Excel (.XLS) version looks like, with conditional cell colors.

• Sample 1 is a qualification run that requires five consecutive measurements in the Green region. As the inspector or operator enters each one, the cell turns green, thus yielding an immediate visual control on the process' performance. Upon entry of the fifth measurement, a green PASS appears in the Results column.
• Sample 2 is a sampling of two measurements. The first is in the Yellow region, and the cell turns yellow as soon as the operator enters it. Since the second result is in the Green zone, PASS appears in the result column.
• Sample 3 also is a set of two measurements but, as both are in the Yellow region, FAIL appears in the result column. It is now necessary to adjust the process and re-qualify it with five consecutive measurements in the Green region.
• Sample 4 is the re-qualification run, but FAIL appears as soon as the first Yellow zone measurement is entered.
• Sample 5 is a successful re-qualification run.
• In Sample 6 (sampling), the first piece is out of specification, and this results in immediate failure.
• Sample 7 is an attempt to re-qualify the process after adjustment, but it fails with the second measurement.
• Sample 8 is a successful re-qualification
• Sample 9 is an example of an unsuccessful qualification run.

The user may ask why it is necessary to put a q in the Type column for qualification runs. The spreadsheet can of course determine whether more than 2 measurements are present (by means of the COUNT function), which has to be a qualification run. Suppose, however, that the first or second measurement is in the Yellow region. The algorithm should immediately tell the operator that a qualification series has failed, and not to make more parts without further action to bring the process back into control. On the other hand, a single Yellow result for a sample of 2 is not a failure.

The important thing to appreciate here is that the conditional cell color provides an immediate visual control on the entry, and one that is probably simpler and more effective than actually plotting the points on a chart. The absence of an actual chart should not be a problem in PRE-Control because trends, run rules, and so on do not apply (as they do in traditional statistical process control). In fact, it is common practice to mark gages (where practical) green, yellow, and red for PRE-Control purposes.