Skip to content
Close
SEARCH
en
de
Contact us
en
de
Contact us
CSP Software Solutions
Worker Guidance
Process Data Management
Database Archiving
Quality Assurance
Anomaly Detection
WHITEPAPER Always audit-ready.  In minutes instead of weeks. Learn how to effectively manage traceability and product liability, immediately limit the scope of an issue in the event of a complaint, and pass audits without having to launch a special project.
Solutions by Industry
Automotive
Aerospace
Mechanical Engineering
Medical Technology
Railway
WHITEPAPER Answers in minutes. Even for records from 2008. Learn how to meet retention obligations without keeping legacy systems alive, erase personal data on time and with documented proof, and answer any regulatory request without turning it into a project.
Knowledge & Practice
Whitepapers, Guides & Success Stories
Blog & News
Events
FREE CHECKLIST How process-reliable is your line?

CPK Calculator for Mass Production

Enter measured values, set tolerance limits—you’ll receive Cp, Cpu, Cpl, and Cpk values with evaluations according to IATF 16949 (1.33) and AIAG-VDA for special characteristics (1.67).

The calculation runs entirely in the browser. Your measurement data never leaves your device.

IATF 16949
Cpk ≥ 1.33
Base value for normal characteristics
AIAG-VDA, VDI/VDE 2645
Cpk ≥ 1.67
Safety and critical characteristics

Measurements

Click a cell and type a value. To fill whole rows, copy from Excel and paste into a cell. At least 2 values, 30 or more recommended.

 
 
Filled: 0
 

Specification limits

 
Cpk — actual process capability
—
Waiting for input

Enter at least two measurements and one spec limit. The result updates automatically.

Cp
—
Cpu
—
Cpl
—

Sample statistics

n
0
Mean x̄
—
σ (n−1)
—
Minimum
—
Maximum
—
Out of spec — observed
—
Expected (normal distribution)
—

Here's how to calculate it

A quick reference guide for understanding the process—identical to the Excel template that many quality departments use anyway.

Formulas

Cp(OGW − UGW) / 6σ
Cpu(OGW − x̄) / 3σ
Cpl(x̄ − UGW) / 3σ
Cpkmin(Cpu, Cpl)

σ is the standard deviation of the (n−1) sample across all values. Strictly speaking, this corresponds to the Ppk methodology; the Cpk calculation within subgroups (X̄-R chart) is not included.

Evaluation according to IATF 16949 and AIAG-VDA

< 1.00 Not capable — parts are produced outside tolerance. Immediate corrective actions, sorting, root cause analysis.
1.00 – 1.33 Borderline — below the IATF baseline value. Improve before PPF/PPAP submission.
1.33 – 1.67 IATF baseline value for normal characteristics — acceptable for submission to most OEMs.
≥ 1.67 Safety and critical characteristics according to AIAG-VDA and customer-specific requirements.

Frequently Asked Questions

How many data points do I need to calculate a reliable Cpk? Mathematically, two data points are sufficient because that is the minimum required to calculate a standard deviation. For a reliable estimate, the AIAG SPC Manual recommends at least 30 data points. For sampling, many OEMs require 30 data points; for special characteristics, they often require 50 or 100. The customer’s requirements and the production control plan are decisive.
What is the difference between Cp and Cpk? Cp compares only the width of the variation with the width of the tolerance and ignores the location of the mean. Cpk takes the location into account. If the process is off-center, Cpk is always less than Cp; the difference is a direct measure of the off-center deviation.
Why does it say Cpk here, even though σ is calculated across all values? The calculator uses the total standard deviation of the (n−1) sample. Methodologically, this corresponds to the Ppk. In daily practice with a continuous series of measurements, this is the standard analysis; for true Cpk within subgroups, the subgroup size of an X̄-R or X̄-s chart is required.
What should you do if the Cpk is too low? First, check the measurement system; without suitable measurement equipment (MSA, Gauge R&R), it is not possible to draw conclusions about the process. Next: if the process is out of control, correct the process status; if the variation is too great, use the Ishikawa diagram and the 5 Whys to identify the dominant source of variation. If there is a risk of nonconforming parts being shipped, implement immediate corrective actions.
Will my measurement data be transmitted? No. The entire calculation runs as JavaScript in the browser. No values are sent to a server, and nothing is stored.

Cpk measures the process.
We secure the underlying database.

The key metric is just the beginning. CSP integrates measurement data, inspection plans, and traceability into a system that detects deviations before they reach the customer.