Question 1

What is Cpk?

Accepted Answer

Cpk (Process Capability Index) measures how well a process produces output within specification limits. It accounts for both the spread and centering of the process. A Cpk of 1.33 or higher is generally considered capable.

Question 2

What is a good Cpk value?

Accepted Answer

A Cpk ≥ 1.33 indicates a capable process (fewer than 63 defects per million). Cpk ≥ 1.67 is considered excellent. Cpk < 1.0 means the process does not meet specifications.

Question 3

What is the difference between Cpk and Ppk?

Accepted Answer

Cpk uses sample standard deviation (n−1 divisor) and reflects short-term process capability. Ppk uses overall (population) standard deviation (n divisor) and reflects long-term process performance. In practice with subgroup data collected over time, Ppk is typically lower than Cpk due to additional between-subgroup variation.

Question 4

What if my data is not normally distributed?

Accepted Answer

Traditional Cpk assumes normal distribution. When data is skewed or multimodal, the standard formula can be misleading. Entropy-based methods like those used by EntropyStat work with any distribution shape, providing more reliable capability estimates.

Free Cpk Calculator