Cp, Cpk, Pp, Ppk, sigma level and estimated PPM. Deterministic, 100% client-side, no upload.
From ungrouped data, Pp/Ppk use the overall sample σ (n−1) and equal Cp/Cpk here. PPM assumes a normal distribution. Cpk ≥ 1.33 capable · 1.0–1.33 marginal · < 1.0 not capable.
Statistical process control requires continuous evaluation of whether manufacturing lines can meet customer specifications. A process capability calculator computes critical indices like Cp, Cpk, Pp, and Ppk from sample data or known mean and standard deviation. By estimating Sigma levels and Parts Per Million (PPM) defect rates, quality engineers can proactively adjust machinery before generating nonconforming scrap.
Cp measures the potential capability of a process if it were perfectly centered between specification limits. Cpk measures the actual capability, penalizing the score if the process mean shifts closer to either specification limit.
Cpk uses short-term standard deviation to measure process potential (what the process is capable of doing). Ppk uses long-term standard deviation to measure actual process performance over time, including external variations.
A Cpk of 1.0 indicates the process mean is exactly 3 standard deviations from the nearest specification limit. The industry benchmark for a capable process is a minimum Cpk of 1.33 (4 sigma), while critical automotive or aerospace parts often require 1.67 (5 sigma) or 2.0 (6 sigma).
PPM translates abstract capability scores into tangible defect rates. For example, a Cpk of 1.0 implies a theoretical defect rate of approximately 2,700 parts per million, allowing management to estimate scrap costs.