To ensure that your results are valid, consider the following guidelines when you collect data, perform the analysis, and interpret your results.
- You must be able to count the number of defects on each item or unit
- If your data are not the total number of defects for each unit in a subgroup, then you cannot estimate the capability of your process using the Poisson distribution. If you can determine only whether each unit is defective or nondefective, use Binomial Capability
Analysis to evaluate the percentage of defective items.
- 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. A subgroup can be either a single unit or a collection of similarly sized units. For example, you could record the number of surface imperfections for one LCD panel (a single unit) or for a collection of LCD panels that are all the same size. If the subgroup is a collection of units, you should collect them under the same inputs and conditions, such as personnel, equipment, suppliers, or environment.
- 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 defect rate per unit times the subgroup size should be at least 0.5. 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
- A subgroup can be either a single unit or a collection of similarly sized units. In either case, the subgroup size can vary and may be defined by a length of time, an area, or a number of items.
- 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 U chart from the Poisson 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.