To ensure that your results are valid, consider the following guidelines when you collect data, perform the analysis, and interpret your results.
- The data must be counts of defective items
- Each item in a subgroup must be classified as either acceptable or not acceptable (defective). If your data are not counts of defectives, you cannot estimate the capability of your process using the binomial distribution. If you have counts of the number of defects on each item, use Poisson Capability
Analysis to evaluate the defects per unit.
- Collect data in subgroups
-
A subgroup is a collection of similar items that are representative of the output from the process you want to evaluate. The items in each subgroup should be collected under the same process conditions, such as personnel, equipment, suppliers, or environment. If you do not collect data in subgroups under the same process conditions, the variation in the subgroups may reflect special causes rather than the natural, inherent variation of the process.
- Collect enough subgroups to obtain reliable estimates of process capability
- Try to collect at least 25 subgroups. If you do not collect a sufficient amount of data over a long enough period of time, the data may not accurately represent different sources of process variation and the estimates may not indicate the true capability of your process.
- The subgroups must be large enough
- The average proportion of defective times the subgroup size should be at least 0.5 for all subgroups. If the subgroup size is not large enough, the control limits may not be reliable when they are estimated from the data.
- The subgroups sizes can be unequal
-
Subgroups can vary in size. For example, if a call center tracks 100 incoming calls each hour and counts the number of unsatisfactory wait times, then all the subgroup sizes are 100. However, if the call center tracks all the incoming calls during a randomly selected hour of the day, then the number of calls is likely to vary and cause the subgroup sizes to be unequal.
- The process must be stable and in control
- If the current process is not stable, then the capability indices cannot be reliably used to assess the future, ongoing capability of the process. Use the P chart from the binomial capability analysis output to determine whether the process is stable and in control. Investigate out-of-control points and eliminate any special-cause variation in your process before you evaluate the process capability.
