What this is
What is a process capability study?
What is a process capability study?
A process capability study is a structured measurement of whether a process can consistently hold its specification, using data collected from a run that has already been shown to be stable. It confirms preconditions, calculates Cp, Cpk and Ppk against the specification limits, and concludes with a decision on whether the process, the specification or the inspection regime needs to change.
Who carries out a process capability study?
Quality engineering, working from data generated on the line or asset being studied. It depends on the measurement system already being validated and the process already being in statistical control, so it cannot be done from the desk without that groundwork.
What does the Cpk figure actually tell you?
Cpk is the capability of the process as it actually runs, accounting for how far the mean sits from centre. A Cpk below 1.00 means out-of-specification output will occur even though nothing has gone wrong; it is simply what the process does. It tells you whether to act on the process or the specification, not whether to inspect more.
Scope
When is a process capability study required?
This assessment is one step in a larger programme. Using it for work that belongs to a neighbouring template produces records that are hard to report on later.
Use this template when
- A new process, line or asset is being qualified before production
- An existing process has changed materially and its capability needs re-establishing
- A customer requirement or a capability concern raises the question directly
- You are running the HACCP and Food Safety or Statistical Process Control programme and this is one of its steps
- A linked record needs this one to exist: an asset qualification or a control plan update depends on the result
Do not use it for
- Control Plan, which sets out what is measured, how often and what to do out of limits, once capability is already established
- Statistical Process Control Log, which records measurements over time rather than summarising a single study
- Setup Approval Record, which confirms a new setup before the run continues past first pieces, with no capability calculation needed
- Gage R&R Study, which measures the measurement system itself and should be run, and referenced here, before capability is trusted
- Anything outside KnowQuality, which belongs in the workspace that owns that process
Compliance mapping
Which ISO 22514 requirements does this satisfy?
ISO 22514 sets out how capability and performance are calculated and what must be true before they mean anything. It is a statistical methods standard, not a management system clause, so its requirements sit inside the study, not around it.
| Clause | Requirement | Where it lands |
|---|---|---|
| ISO 22514-1 | General principles and vocabulary for capability studies, including that the process be assessed for stability first | Preconditions |
| ISO 22514-1 | The measurement system used to generate the data is fit for the study's purpose | Preconditions |
| ISO 22514-2 | Capability and performance measures for a normally distributed characteristic, covering Cp, Cpk, Pp and Ppk | Results |
| ISO 22514-2 | The distributional assumption behind the normal-based indices is checked, not assumed | Preconditions |
| ISO 22514-2 | Distinction between potential capability, unaffected by centring, and actual capability, which accounts for it | Results |
| ISO 9001 cl.9.1.1 | Monitoring and measurement of process performance, of which this study is one instrument | Results |
| Codex Alimentarius HACCP, Principle 6 | Verification that confirms the system works as intended, which a study on a controlled characteristic can support | Conclusion |
What it does not cover
- Gage R&R Study, which quantifies how much observed variation comes from the measurement system, and should be run before a capability figure is trusted.
- Statistical Process Control Log, which tracks a characteristic over time on a control chart, rather than summarising one study.
- Control Plan, the standing document a capability study updates, not a substitute for one.
- Setup Approval Record, a first-piece signoff that does not require a capability calculation at all.
- HACCP Verification Record, which documents the wider verification activity a study may feed into, but is not itself a capability instrument.
Global
Process Capability Study requirements by country
ISO 22514 is a voluntary standard, not a legal instrument, so what varies by jurisdiction is less the duty to study capability and more who expects to see the evidence, and in what form.
ISO 22514 (parts 1 and 2)
Defines how capability and performance indices are calculated and what has to be true of the data first.
Certification auditors ask to see the preconditions checked, not just the final Cpk.
Customer quality agreements; FDA current good manufacturing practice for food processing
No standalone federal requirement to run a capability study, but customer specs and food safety plans routinely require one at qualification.
The obligation usually arrives through a contract or a HACCP plan, not a statute.
BRCGS Food Safety standard; customer technical agreements
Certification schemes commonly require capability evidence for critical control points and key characteristics.
An auditor expects the study to be traceable to the characteristic it was run to protect.
How to complete it
How to complete a process capability study, step by step
The template prompts for the numbers. What decides whether the study is trustworthy comes before and after them.
Confirm statistical control and a validated measurement system first, and record how each was confirmed rather than simply selecting Yes. A study built on an unconfirmed precondition is not a capability study at all.
Cp shows what the process could achieve if perfectly centred; Cpk shows what it actually achieves. A large gap means the fix is centring, usually faster and cheaper than reducing variation, and the study should say so explicitly rather than leaving the reader to infer it.
Cp and Cpk assume a normal distribution. Skewed or bounded data needs a transform, or a different index, and the study should record which was used. An index on the wrong assumption misleads in either direction.
A not capable rating should not conclude in continue monitoring, nor a capable one in immediate action. Where the recommendation and the rating disagree, that disagreement is itself the finding, and belongs in the record rather than being resolved silently before submission.
What auditors find
Most common process capability study findings
The template almost always contains a number. The findings concern whether it was earned and whether anything downstream responded to it.
| Finding | Clause | What fixes it |
|---|---|---|
| Capability calculated before statistical control was confirmed. | ISO 22514-1 | Require the control check to pass before the results section can be completed. |
| Cpk reported with no evidenced measurement system validation. | ISO 22514-1 | Link the Gauge Study ID and confirm the validation date precedes the data collection period. |
| Data not checked for normal distribution, or found non-normal with no transform recorded. | ISO 22514-2 | Record the distribution check and the transform used, or the alternative method applied. |
| Recommended action does not follow the capability rating, e.g. continue monitoring against a not capable result. | ISO 22514-2 | Gate the recommended action options against the rating so an inconsistent pairing is visible before closure. |
| A wide gap between Cp and Cpk is recorded but not addressed as a centring problem. | ISO 22514-2 | Require a stated diagnosis, centring or variation, whenever the gap exceeds a defined threshold. |
| Action required is marked yes with no CAPA reference captured. | Standard-specific | Make the CAPA ID a condition of closing the study once action required is set to yes. |
Case in point
Case in point: the study that was capable and still wrong
A packing line ran a capability study on fill weight after a filler change. Cp came back at 1.58 and Cpk at 1.31, both comfortably capable, and the study concluded continue monitoring with no further action. The gap between Cp and Cpk, roughly a fifth of potential capability lost to centring, sat in the figures uncommented on in the conclusion.
Three months later the mean had drifted further off centre, in the same direction the original gap already pointed to, and underweight complaints arrived from a retail customer. The reassessment found the process's inherent variation unchanged; it was still capable in the Cp sense. What had changed was centring, which the first study had already flagged as the weaker figure and which nobody had asked the line to correct, because the recommendation had not required it.
The template
The template, field by field
The form exactly as it installs. Every field, option, score and conditional rule is editable, and the links to other templates come with it.
4 sections
- Reference
- QUA-059
- Archetype
- Assessment
- Record ID
- CAP-2026-000
- Scoring
- Cpk value
- Direction
- High is good
- Singleton
- No
- Basis
- ISO 22514
- Links
- Links Asset, Control Plan
- Tags
- Process, Analysis
- Sections
- 4
- Fields
- 42
- Follow up fields
- 3
- Repeating sections
- 0
- Links out
- 5
Header
13 fieldsStudy ID*
Auto sequence. Format CAP-2026-0000.
The record's own ID. Other templates point at this value.
Status*
Drives who this goes to next.
- Planned2 pts
- In progress2 pts
- Complete3 pts
- Deferred0 pts
- Open0 pts
- Closed3 pts
- Overdue0 pts
Date and Time*
Raised By*
Site*
Site ID*
Format SITE-000.
Links to FDN-001 Site ID
Line*
Shift*
Line Lead
Characteristic Studied*
Study Trigger*
Asset
Asset ID
Format AST-0000.
Links to FDN-002 Asset ID
Preconditions
7 fieldsStability Comes First
Capability calculated on an unstable process is meaningless. Confirm statistical control before you calculate anything.
Process In Statistical Control*
- Yes3 pts
- Marginal1 pt
- No0 pts
Measurement System Validated*
- Yes3 pts
- No0 pts
Gauge Study ID
Links to QUA-066 Study ID
Data Points Collected*
Period Covered*
Data Normally Distributed*
- Yes3 pts
- Transformed2 pts
- No0 pts
Results
9 fieldsSpecification Limits*
Process Mean*
Standard Deviation*
Cp
Potential capability if the process were perfectly centred.
Cpk*
Actual capability accounting for centring. This is the number that matters.
Ppk
Capability Rating*
- Excellent, Cpk above 1.674 pts
- Capable, Cpk 1.33 to 1.673 pts
- Marginal, Cpk 1.00 to 1.331 pt
- Not capable, Cpk below 1.000 pts
Process Centred*
A large gap between Cp and Cpk means the process is capable but off centre, which is usually the easier fix.
- Yes3 pts
- Slightly off1 pt
- Clearly off0 pts
Estimated Defect Rate PPM
Conclusion
13 fieldsProcess Capable*
- Yes3 pts
- Marginal1 pt
- No0 pts
Recommended Action*
Centre the process, reduce variation, widen the specification or increase inspection.
- Continue monitoring3 pts
- Increase monitoring frequency2 pts
- Plan corrective work1 pt
- Immediate action0 pts
Inspection Frequency Change*
A capable process earns reduced inspection. An incapable one needs more.
- Reduce3 pts
- No change2 pts
- Increase1 pt
Control Plan Update Required*
Control Plan Reference
Links to QUA-057 Plan ID
Action Required*
Raise the action record, then enter its reference here.
- No2 pts
- Yes0 pts
Priority
- High0 pts
- Medium1 pt
- Low3 pts
CAPA ID
Format CAPA-2026-00000.
Links to FDN-014 CAPA ID
Action Owner
Quality Engineering*
Signature*
Production Manager*
Second Signature*
QUA-059 · record IDs look like CAP-2026-000 · Links Asset, Control Plan
Open in KnowellaRun it with agents
From a document you fill in to a programme that runs itself
The study is a snapshot. What decides whether it changes anything is whether the linked asset, control plan and action records actually move once the number is in.
Holds the capability study library against the asset and control plan registers, flags studies whose linked asset has since changed, and routes the result to the control plan owner.
Surfaces the asset's capability history alongside its maintenance record, so a declining Cpk trend is visible before the next study is due.
Tracks the gauge study behind the measurement system validation and flags a study whose gauge evidence has expired.
Coordinates the crew, rolls capability results and open actions into one view, and holds every write for approval before it touches a record.
This template lives in KnowQuality — quality and food safety. HACCP, nonconformance, traceability, laboratory and customer complaints.
Meet KnowQuality→Glossary
Process Capability Study definitions and key terms
- Cp
- Potential process capability: what the process could achieve if perfectly centred between the specification limits, ignoring where the mean actually sits.
- Cpk
- Actual process capability, accounting for how far the mean is from centre. The lower of the two one-sided capability figures, and the number that describes what the process delivers today.
- Ppk
- Performance index calculated like Cpk but using overall rather than within-subgroup standard deviation, capturing longer-term variation a shorter study may not see.
- Statistical control
- The state in which a process shows only the variation inherent to it, with no special causes present, and the precondition every capability index depends on.
- Estimated defect rate (PPM)
- The parts-per-million rate implied by the capability figure and the specification limits, translating the index into an expected count of nonconforming output.
FAQ
Frequently asked questions about process capability study
What is a process capability study based on?+
It is built against ISO 22514, the international standard for capability and performance measures, which sets out how Cp, Cpk, Pp and Ppk are calculated and what has to be confirmed about the data before they are used.
Why does the template ask about statistical control before it asks for the specification limits?+
Because a capability index calculated on a process that is not in statistical control is not describing the process's ordinary behaviour; it is describing a mixture of ordinary variation and something else, and the resulting number is not usable regardless of how it is presented.
What should we do when Cp is good but Cpk is poor?+
Treat it as a centring problem rather than a variation problem. The process has the inherent capability to meet the specification; it is simply running off-centre, and adjusting the target or setup is usually faster and cheaper than trying to tighten the spread of the process itself.
Does a capable result mean inspection can be reduced immediately?+
Only if the study also confirms the process is stable and the measurement system trustworthy. A capable Cpk on data that has not cleared those preconditions is no basis for reducing inspection.
How does this template connect to the Control Plan?+
A capability result that changes the picture should prompt a Control Plan update: changed inspection frequency, a changed control method, or a note that the specification needs review. The link exists so a study does not sit unread once filed.
Which programmes use this template?+
Both HACCP and Food Safety, where it can support verification of a controlled characteristic, and Statistical Process Control, alongside the SPC log and the gauge study it depends on.
Keep going
Related templates and programmes
Industries this is written for
Programmes this belongs to
Used together in HACCP and Food Safety
Unannounced Audit Readiness Check
Checks whether the site would pass an audit arriving today, covering records, standards and area condition
Product Nonconformance Report
Records product that does not meet specification
Hold and Release Record
Places suspect product on hold and records the decision to release, rework or scrap it
HACCP Plan
The hazard analysis and critical control point plan for a product or process
HACCP Verification Record
Confirms the HACCP plan is being followed and still works
CCP Monitoring Log
Records readings at a critical control point, such as cook temperature, chill time or metal detection
More in Process Control
Control Plan
Sets out what is measured on a process, how often, by what method and what to do when it goes out of limits
Statistical Process Control Log
Records measurements over time so process variation can be seen and acted on before product fails
Setup Approval Record
Confirms a new setup is correct and approved before the run continues past the first pieces
Batch Record
Records everything that happened during the production of one batch, including materials, settings, checks and people
Yield Analysis Record
Compares material in against product out and accounts for the difference
Shift Handover Record
Passes on what the incoming shift needs to know, including running issues, holds and changes
Written and reviewed by
Siddarth Singh
Founder & Chief Executive Officer, Knowella
Certified Safety Professional and industrial and systems engineer with more than a decade inside food supply chain, freight and manufacturing operations. This page was written against the current text of the standards it cites, not against secondary summaries of them.
- Certified Safety Professional (CSP), Board of Certified Safety Professionals
- MBA, University of Chicago Booth School of Business
- MS and BS, The Ohio State University, Industrial and Systems Engineering
- Six Sigma Black Belt
Sources and last review. Reviewed 16 August 2026 against:
- ISO 22514-1 — Statistical methods in process management, capability and performance, general principles and concepts
- ISO 22514-2 — Process capability and performance of processes following a normal distribution
- ISO 9001:2015 clause 9.1.1, monitoring, measurement, analysis and evaluation
- Codex Alimentarius General Principles of Food Hygiene, HACCP annex, Principle 6
This page is general guidance, not legal advice. Confirm requirements with your jurisdiction’s regulator.