A comprehensive process capability analysis tool that contains an executive summary for communicating performance measures to project stakeholders, detailed capability reports, and diagnostic tools for verifying the validity of your assessments. This report is available only when the Six Sigma add-on module for Minitab is installed.

Answers the questions:

- What is the capability of the process (both long-term and short-term) at the start of the process improvement project?
- What is the capability of the process (both long-term and short-term) after improvements have been made?
- Was the process stable during these assessments?
- Did you collect enough data to trust the assessments?

When to Use | Purpose |
---|---|

Start of project | Perform a baseline capability analysis on the process to determine its performance at the start of the project. A baseline analysis helps you set improvement goals for the project. |

Mid-project | Perform a confirmation capability analysis after improvements have been implemented to confirm that the process performs as predicted. |

End of project | Perform a capability analysis after implementing controls to obtain a final assessment of process capability, and also to determine whether the improvement goals of the project were attained. |

Continuous Y (output), at least one specification.

- Verify the measurement system for the Y data is adequate.
- Establish a data collection strategy to define how you will sample subgroups over time. Ensure that you are using rational subgroups whenever possible.
- Collect data for the rational subgroups and enter them into Minitab. In the Minitab worksheet, you can enter all the data in a single column or you can enter each subgroup into a separate row. Minitab can also directly import from databases, text files, Excel, and so on.
- You must provide at least one specification limit to produce the process report.
- Specification limits can be defined as a boundary. When you use a boundary, you are indicating that it is impossible to have data outside of the specification limit (for example, the yield from a chemical process cannot be greater than 100%). In that case, Minitab does not calculate the expected DPMO for whichever limit has been defined as a boundary. Note: If you define both specification limits as boundaries, Minitab does not calculate any expected DPMO because you have said it is impossible to have a defect).
- You may choose to use the Box-Cox transformation if your data are not normal.

- Data are assumed to come from a normal distribution; however, the consequences of having nonnormal data are not serious if the data are reasonably normal. Badly skewed data can often be easily fixed using a transformation, such as the Box-Cox.
- The short-term Z statistics in the Six Sigma Process Report are different from those reported in other Minitab process capability reports because the Six Sigma Process Report uses an absolute best-case approach (the process is centered on its target and has only common-cause variation) when calculating the short-term Z, while the Minitab's capability reports use a less-than-best-case approach (the process has only common-cause variation, but it is assumed to be centered on its mean instead of its target). The Six Sigma Process Report, therefore, gives a better estimate of the absolute process potential.
- Sometimes, you may find collecting data in rational subgroups difficult. In these cases, you should use the long-term standard deviation and resulting long-term Z to establish the short-term Z (by subtracting off a typical value for process shift) than to estimate the short-term Z from a moving average. Using the moving average requires that consecutive items are similar, but this assumption may not be valid in many cases.
- Use the cumulative graphs in the Six Sigma Process Report to provide evidence that you have collected enough data. If these graphs stabilize (plateau), you have collected enough data, and the process is relatively stable. If these graphs have not stabilized, either you did not collect enough data, or the process is not stable. The control charts in the report can help you determine whether the process is stable.
- If you have discrete numeric data from which you can obtain every equally spaced value and you have measured at least 10 possible values, you can evaluate these data as if they are continuous.