Quality & InspectionFree calculatorPDF, Excel, Word exportSAP S/4HANA

Cp and Cpk Process Capability Calculator for SAP QM Results Recording

Cp and Cpk summarise how well a process fits its tolerance. Customers in automotive, aerospace and medical ask for them in PPAP and supplier audits. If you record results against characteristics in SAP QM, you can pull the mean and sigma and test capability here.

Your numbers

Upper specification limit

Lower specification limit

Process mean

Average of recorded characteristic values.

Process standard deviation

Use the within-subgroup sigma for Cp/Cpk, overall sigma for Pp/Ppk.

Your results

Cpk

1.1111

Expected PPM out of spec

429

1.3889

Potential capability, ignores centring.

Cpu

1.1111

Cpl

1.6667

PPM above USL

429

PPM below LSL

Expected yield

100%

Assumes a stable, normally distributed process. Check stability on a control chart and normality before quoting indices to a customer.

Save results:

Cpk as the mean drifts across the tolerance

Sweeps Process mean from 9.95 to 10.05, holding your other inputs fixed.

Capability reporting from your ERP

Optional: automate capability indices across parts and vendors from SAP QM results.

Optional. The tool and every export work without this. No spam.

How to use it

Enter the specification limits from the master inspection characteristic, then the mean and standard deviation of your recorded values. The tool reports Cp, Cpk and the expected PPM beyond each limit using the normal distribution. The chart shows Cpk as the mean moves from LSL to USL, which makes the cost of poor centring obvious: a process with Cp 2.0 can still have a low Cpk if it is off centre.

Formulas

Cp = (USL - LSL) / 6 sigma. Cpu = (USL - mean) / 3 sigma and Cpl = (mean - LSL) / 3 sigma. Cpk is the smaller of the two. Expected PPM is the normal tail beyond each limit, scaled by one million. A centred process with Cpk 1.33 gives about 63 PPM across both tails, and 1.67 is a common target for critical characteristics.

Cpk below 1.00: process not capable
1.00 to 1.33: marginal
1.33 to 1.67: capable
Above 1.67: highly capable

Data quality matters

Sigma must come from a stable process and enough data, ideally 100 or more individual readings in rational subgroups. A sigma computed from a mixed population of machines or lots understates capability problems. Also confirm that the gauge is adequate first; a poor measurement system inflates sigma and makes a good process look bad.

Common mistakes

Using the range of the specification instead of the sigma, mixing short-term and long-term sigma, and applying the normal model to a bounded characteristic such as flatness or runout. For one-sided specs, only compute the index for the limit that exists.

Frequently Asked Questions

What is the difference between Cpk and Ppk?

Cpk uses within-subgroup variation and describes potential capability when stable. Ppk uses overall variation including shifts between subgroups. Enter the matching sigma to get either.

What Cpk do customers usually require?

Many automotive customers ask for at least 1.33 on ongoing production and 1.67 on initial studies, but your customer specific requirements decide.

Can I use a one-sided specification?

Yes, set the missing limit far away so its tail contributes essentially zero, then read Cpu or Cpl directly.

Why is Cpk lower than Cp?

The mean is off centre. The gap between Cp and Cpk is a quick indicator of how much capability you could regain by centring the process.

We can help extract result values from SAP QM and compute capability across materials and vendors automatically, with the query shown.

Talk to an Expert Explore More Free Tools

Related Tools

Quality & Inspection

More Quality & Inspection tools

See all

Back to All Tools