Find the Root Cause Behind Process Variation and Yield Loss
CPK shows there is a problem — but not where it starts. QualityLine helps you identify the exact source of variation across your production process.
What Is Cpk (Process Capability Index)?
A statistical measure used in manufacturing to quantify how well a process performs relative to its specification limits — taking both variation and centering into account.
Cp — Process Capability
Cp measures how much of the specification range your process could fit if it were perfectly centered. It evaluates the spread of your process relative to the tolerance window, ignoring whether the process is actually centered within that window.
Cpk — Process Capability Index
Cpk adjusts Cp to account for where the process is centered. It measures the distance from the process mean to the nearest specification limit. A high Cpk means the process is both tight and centered.
What Does Cpk Measure?
Cpk captures four critical dimensions of process performance in a single index value.
Process Centering
How close the process mean is to the center of the specification range.
Process Variation
The spread of output values relative to the tolerance window.
Defect Probability
The likelihood that a unit will fall outside specification limits.
Capability vs Tolerance
Whether the process variation fits within engineering tolerance — and by how much.
Key Insight
A process can have high Cp but low Cpk if the mean is shifted toward one limit. Cpk is more reliable because it reflects actual defect risk.
Cpk Formula
The Cpk formula takes the minimum of two ratios. The smaller value determines process capability.
Upper Specification Limit
The maximum allowable value. Units above this are out-of-spec.
Lower Specification Limit
The minimum allowable value. Units below this are out-of-spec.
Process Mean
The arithmetic average of the measured process output values.
Standard Deviation
A measure of process spread. Smaller σ means tighter output.
Three Sigma Spread
±3σ covers 99.73% of output under a normal distribution.
Minimum Function
Cpk takes the smaller ratio — only as capable as the closest spec limit.
How to Calculate Cpk Step-by-Step
Follow these five steps to calculate Cpk for any manufacturing process.
Define the specification limits
Obtain USL and LSL from the engineering drawing, customer specification, or process control plan.
Collect a representative data sample
Gather at least 30 data points from the process under stable, controlled conditions.
Calculate the process mean (μ)
Sum all measured values and divide by the count.
Calculate the standard deviation (σ)
Use the sample standard deviation formula.
Apply the Cpk formula
Calculate both ratios and select the minimum.
Cpl = (μ − LSL) / 3σ
Cpk = min(Cpu, Cpl)
Not sure what's driving your Cpk drop?
We'll show you where variation starts — using data from your own process.
Cpk Calculation Example in SMT Manufacturing
A practical walkthrough using solder paste volume measurement.
Scenario: SPI Solder Paste Volume
Spec: 80–120% nominal. 50 boards measured. Mean μ = 97%, σ = 6%.
Cpk Calculation — SPI Paste Volume
Step-by-stepDifference Between Cp and Cpk
Both measure process capability — but they answer different questions.
| Dimension | Cp | Cpk |
|---|---|---|
| What it measures | Potential capability assuming perfect centering | Actual capability accounting for mean location |
| Centering sensitivity | NO | YES |
| Formula | (USL−LSL)/6σ | min[(USL−μ)/3σ,(μ−LSL)/3σ] |
| When equal | Only when process mean is exactly centered. Rare in practice. | |
| Which to use | Assessing potential spread | Always for pass/fail assessment |
What Is a Good Cpk Value?
Cpk thresholds define whether a process is acceptable for production.
Free Online Cpk Calculator
Enter your process parameters for instant Cp, Cpk and sigma level results.
Cpk / Cp Calculator
Works for solder paste volume, placement offset, reflow temperature, test voltage, and any measurable process parameter.Download Cpk Benchmark Reference Card
One-page PDF: thresholds, industry standards, and common failure modes for SMT / EMS.
CPK shows the problem — not the cause
Most teams calculate Cpk only after defects have already happened. By the time the number drops below 1.33, the damage is done.
- Scrap already occurred — units are built, inspected, and rejected
- Yield already dropped — FPY numbers reflect last shift, not this one
- Root cause is harder to trace — the signal is buried under hours of production data
The real advantage is identifying variation before it impacts production — not calculating what went wrong after the fact.
Limitations of Manual Cpk Monitoring
Calculating Cpk manually introduces blind spots that make real-time process control impossible.
Static Snapshots
Manual Cpk is calculated at a point in time. Between measurements, the process can drift significantly without detection.
Delayed Detection
Cpk degradation is detected hours or days later — after defective units have already been produced and shipped.
Spreadsheet Dependency
Manual tracking relies on Excel exports and manual interpretation. Time-intensive, error-prone, and impossible to scale.
Multi-Line Blind Spots
Tracking Cpk manually across 8–20 SMT lines is operationally infeasible. Unmonitored lines are where the next crisis starts.
How QualityLine identifies root cause
QualityLine connects data across every stage of the production line — from paste deposition to final test — and applies AI correlation to surface the true source of variation, not just where it appears.
AOI / SPI Inspection Data
Solder paste volume, component placement offset, tombstoning, and missing components — ingested automatically from every inspection station.
Test Systems (ICT / FCT)
In-circuit and functional test results correlated to upstream process parameters in real time — not reviewed manually after the shift.
Machine & Process Parameters
Printer speed, squeegee pressure, reflow profile, placement force — every controllable variable tracked and analyzed continuously.
Repair & Failure Data
Rework records and failure codes linked back to originating process steps — closing the feedback loop between quality and production.
Real-Time Process Capability Monitoring
Modern manufacturing analytics platforms continuously monitor Cpk across all lines and stations — automatically.
Continuous Cpk Tracking
Capability recalculated in real time as each measurement arrives — SPI, AOI, ICT — without manual export.
Automated Cpk Alerts
When Cpk drops below threshold, alerts route automatically to engineers before defects scale.
Root Cause Correlation
Cpk degradation correlated against all data sources to identify root cause without manual analysis.
Multi-Line Visibility
All lines and stations visible in one dashboard. Know which lines are drifting without visiting the floor.
Frequently Asked Questions
Direct answers from manufacturing engineers and quality teams.
See what's causing variation in your process
Tell us a bit about your process — we'll show you where variation typically starts and how it impacts yield.
- 15–20 min technical walkthrough
- See how variation is detected across your process
- Understand what's impacting your yield
