Process Capability Index Calculator

Cp (potential capability) measures how well the process could fit within specification limits if it were perfectly centered, comparing the total tolerance width to the process spread (6σ). Cpk (actual capability) accounts for where the process mean actually sits relative to both limits, taking the minimum of the upper and lower capability indices. A process can have a high Cp but a low Cpk if the mean is shifted away from the center of the specification.

A Cpk of 1.33 means the process mean is at least 4 standard deviations away from the nearest specification limit, leaving a safety buffer beyond the standard 3-sigma requirement. Many industries use 1.33 as the minimum acceptable threshold because it accounts for small process shifts over time and reduces the risk of producing defects. For critical processes, a Cpk of 1.67 or higher is often required.

A Cpk below 1.0 indicates the process is not capable of consistently producing output within the specification limits — some products will fall outside the acceptable range. The lower the Cpk, the higher the defect rate. Immediate corrective action is typically needed, either by reducing process variation (lowering σ) or shifting the mean closer to the specification center.

Cpk is always less than or equal to Cp. The gap between them reflects how off-center your process mean is relative to the midpoint of the specification limits. A large difference means the process is well-capable in spread but poorly centered. Adjusting the process mean toward the target midpoint (USL + LSL) / 2 will bring Cpk closer to Cp without changing variation.

Yes. A negative Cpk means the process mean has moved outside one of the specification limits entirely, indicating the process is producing predominantly out-of-spec output. For example, if the mean exceeds the USL, the CPU value will be negative, which becomes the minimum and therefore the Cpk. This is a critical situation requiring immediate process correction.

Cpk uses the within-subgroup (short-term) standard deviation and represents the potential performance of a stable process. Ppk uses the overall (long-term) standard deviation calculated from all individual measurements and represents actual long-term process performance. Cpk is typically higher than Ppk; the gap between them indicates how much variation is introduced by process shifts or special causes over time.

Most quality standards recommend a minimum of 30 individual measurements to estimate the standard deviation with reasonable confidence. For subgroup-based studies, collecting at least 20–25 subgroups of size 4–5 is common practice. Smaller sample sizes produce less stable standard deviation estimates, making Cpk values less reliable and potentially misleading.

Cpk calculations assume the process data follows a normal (Gaussian) distribution and that the process is in a state of statistical control (stable over time). If the data is non-normal or the process is unstable with special causes present, Cpk values can be misleading. Always verify process stability with a control chart before interpreting Cpk results.