← Back to blog

Control Plan Format: Columns, Phases and a Worked Example

November 20, 2026
Control Plan Format: Columns, Phases and a Worked Example

A control plan format is a table that links each process step to the characteristic you control, the specification, the measurement method, the sample size and frequency, the control method and the reaction plan. Build it from your PFMEA, fill in one row per step and characteristic, and keep it matched to what operators and inspectors actually do on the floor. This guide gives you the fields, the three phases, a worked example you can adapt and the checks that keep the plan audit-ready. Inspection results then feed straight back into the plan, which is where QA-Report fits in.

QA-Report
Connect Inspection Results to Your Plan
Create dimensional inspection reports that link drawing dimensions to measured results, helping keep inspection records aligned with your control plan.
Visit

TL;DR

  • A control plan lists, for every process step, what is controlled, how it is measured, how often, with what control method and what happens when something goes wrong.
  • Start from your process flow diagram and PFMEA. Put in the key risks and special characteristics, not every conceivable check.
  • Use three versions over the product life cycle: prototype, pre-launch and production.
  • The most common audit finding is a plan that does not match actual practice, so connect plan, inspection records and revisions.
  • Since March 2024, the AIAG Control Plan Reference Manual is the current US automotive reference, subject to customer-specific requirements.

Table of Contents

What a control plan is, and what it is not

AIAG's description is a written summary of the systems used to minimize process and product variation. It covers the actions needed at each process step, including receiving, in-process, outgoing and periodic requirements, so that outputs stay in a state of control. AIAG's control plan guidance also states that it is maintained through the product life cycle and does not replace detailed operator instructions.

That last point shapes the format. The control plan is a summary document. It points to the work instruction, the gauge procedure or the inspection plan by document ID instead of reproducing them. If you are unsure where the boundary sits, our comparison of a control plan and an inspection plan draws it clearly.

In aerospace, AS9145 (the APQP standard for the sector) describes the control plan as a documented description linking manufacturing process steps to key inspection and control activities, according to NQA's APQP control plan overview. Automotive, aerospace and medical device customers each add their own expectations, but the backbone of the table is the same.

What a control plan is, and what it is not

The control plan format: fields to include

There is no single mandatory layout. Practitioners at CQE Academy argue that the format varies widely between industries and companies, and that the content is what matters. Still, auditors look for the same information, and the AIAG form groups it in a recognizable order. A practical layout has four blocks.

Header: general data

The header identifies the plan and its scope. Include:

  • Control plan number and revision, with issue and revision dates
  • Customer and organization name and site
  • Part number, part name and engineering change level
  • Phase covered: prototype, pre-launch or production
  • Key contact and approvals

The IATF 16949 clause 8.5.1.1 requirements, summarized in this IATF 16949 control plan walkthrough, cover this same general data, so a header built this way is a safe default. If you want the surrounding certification requirements, see our IATF 16949 requirements guide.

Process columns

Each row ties to a step in your process flow diagram:

  1. Part or process step number, matching the PFD and PFMEA numbering
  2. Process name or operation description
  3. Machine, device, jig, fixture or tool

Characteristics and specification

  1. Product characteristic (what the part must have, such as a bore diameter)
  2. Process characteristic (what the process must hold, such as spindle speed or clamp pressure)
  3. Special or key characteristic classification
  4. Specification or tolerance

In aerospace, NQA's field list splits classification into Key Characteristics, where variation significantly affects fit, performance, service life or producibility, and Critical Items, which have a significant effect on safety, performance, form, fit or function. Match the classification terms to your customer's flowdown, not to a generic template.

Methods and reaction

  1. Evaluation or measurement technique
  2. Sample size
  3. Sample frequency
  4. Control method
  5. Reaction plan

These five columns are where plans succeed or fail, so they get their own section below.

Prototype, pre-launch and production plans

The plan is not a single document. AIAG frames it as an output of APQP at three stages: a prototype build control plan, a pre-launch control plan and a production control plan, as described in the AIAG material on the control plan as a sixth quality tool. IATF 16949 likewise expects coverage of prototype, pre-launch and production phases, as Advisera's IATF 16949 control plan guide explains.

PhaseTypical contentWhat changes next
PrototypeDimensional and material checks and performance tests on developmental partsLearnings flow into process design
Pre-launchControls on off-tool parts, with added checks and tighter sampling while the process is provenFrequencies relax as capability is demonstrated
ProductionAll product and process special characteristics, process controls and measurement methods for regular productionReviewed after every change or nonconformance

The phase descriptions follow ProjectManager's breakdown of control plan levels. The 2024 AIAG manual adds a Safe Launch requirement to the phases, according to the AIAG manual overview.

Tier suppliers in automotive sometimes work with a generic control plan, which combines process controls by manufacturing process, or a specific plan, which details controls by part number. The choice depends on customer arrangements and product family complexity, as Qualitywise describes. Ask your customer which they accept before you build forty near-identical plans.

For how these plans fit into the larger program, read our guide to the five APQP phases.

Writing the columns that matter

Start from the PFD and PFMEA

Complete the process flow diagram and PFMEA first, so every step and quality risk is accounted for. controlplan.org's how-to guide follows that sequence: list steps in order using the same numbers as the PFD and PFMEA, flag critical-to-quality characteristics, then fill in the control columns. The control plan is risk-based: the PFMEA ranking decides the level of control, and the controls you choose should justify the detection rating, per CQE Academy.

Keep the scope disciplined. NQA notes that a common misconception is that control plans must be complex and include every identified risk. Focus on the key risks from the PFMEA and the plan stays short enough to use. If your PFMEA still lives in a spreadsheet nobody revises, our PFMEA software guide covers the options.

Measurement technique

Name the gauge or method, not just "measure". Before you rely on a measurement system listed here, run a measurement systems analysis covering repeatability, reproducibility, linearity, stability and accuracy, as the IATF 16949 walkthrough points out. A tolerance of ±0.05 mm checked with a gauge that cannot resolve it is a finding waiting to happen.

Sample size and frequency

Sample size is the subgroup you pull: a value of 5 means five parts are measured, a common choice for control charts. Frequency is how often you pull that subgroup. In the AIAG-based material presented at an AIAG event session, if inspection is not 100%, the frequency should be volume-based and support containment. "Once per shift" works poorly when a shift produces 40 parts one day and 4,000 the next. "Every 250 pieces" does not.

Control method

The control method is the mechanism that keeps the characteristic in control. Examples include X-bar and R charts, Np charts, pre-control charts, checklists, log sheets, mistake-proofing and first-piece inspection. Error-proofing is generally more effective than sampling inspection, so list it first where the process allows it. Reference the detailed procedure by document ID instead of pasting it into the cell.

Reaction plan

A reaction plan tells the operator what to do the moment a check fails. AIAG's clarified expectation covers three things: containing suspect product, stopping the process from making more, and bringing the process back into control. NQA adds that it should include segregation of nonconforming product, a correction method and contact with the supervisor, and may include sorting, rework or repair, and customer notification.

Three actions in a control plan reaction

Write reactions an operator can carry out without a meeting. "Notify quality" is weak. "Stop machine, segregate parts since the last good check, call QA" is a reaction plan.

Control plan example with realistic entries

The rows below adapt the sample entries in Indupath's step-by-step control plan guide. Use them to calibrate the level of detail, not as values to copy.

StepCharacteristicSpecificationMeasurementSample size and frequencyControl methodReaction plan
DrillingHole diameter10.00 ± 0.05 mmDigital micrometer1 per hourSPC (X-bar/R)Stop machine, segregate last 10 pcs
TorqueBolt tightening torque25 ± 1 NmTorque wrench100%Poka-yoke and visualInform QA, retighten and recheck
CleaningSurface oil levelClean surfaceVisualEach batchVisual checkReclean and re-inspect

Notice what each row does. The characteristic is specific, the tolerance is numeric, the method names a tool, the frequency is countable and the reaction is an action. A row that fails any of those tests will be questioned at audit.

An aerospace example from AiroQue's published sample plan shows the same logic in a cable assembly. A crimp operation uses 100% visual inspection plus a pull test per lot, and the reaction is to stop production, quarantine every unit since the last passing pull test, verify tool calibration and the die, and raise a nonconformance. Pin depth is measured 100% with a calibrated gauge, with key characteristic results retained per contract.

How the plan stays in sync with the shop floor

The IATF-related findings analysis from simpleQuE reports that most control plan nonconformities occur because the plan does not match actual practice. Auditors check whether the PFD, PFMEA and control plan agree, whether inspection records agree with the plan, whether special characteristics from the part print are included, and whether receiving, warehousing, shipping and outsourced processes are covered.

Treat that as your internal checklist:

  1. Does every special characteristic on the drawing appear as a row?
  2. Does every row's frequency match what the inspection records show?
  3. Do step numbers match the PFD and PFMEA?
  4. Are incoming and outgoing steps included, not only machining?
  5. Is the revision on the plan the one operators actually have?

Review triggers are specific. IATF 16949 expects a review after shipping nonconforming product, after any product or process change, after customer complaints with corrective action, and at a risk-based frequency, per Advisera. Revision discipline matters here, and our revision control implementation plan shows how to keep one live version.

What changed in 2024 and 2025

The AIAG Control Plan Reference Manual, first edition, replaced the earlier APQP and Control Plan second edition manual from March 1, 2024, unless the customer specifies otherwise. Compliance is set by customer-specific requirements. According to simpleQuE, GM and Stellantis set an implementation date of September 1, 2024, and Ford set December 31, 2024.

The IATF Global Oversight Bureau also published updates. GM's customer-specific requirements update, effective January 31, 2025, aligned clause 8.5.1.1 with the current AIAG control plan manual. Ford's update, effective February 6, 2025, clarified control plan expectations for sub-tier critical characteristics and revised Level 3 and 5 PPAP approval requirements. Check your customer's current requirements page before you finalize a format, because these documents are revised periodically.

If PPAP is part of your submission, our PPAP flow diagram checklist shows where the control plan sits in the package.

Where inspection data meets the control plan

A control plan is only as credible as the records behind it. Each row promises a measurement at a given frequency, and the evidence is the inspection report. That is why many quality teams move from scattered spreadsheets to a platform where drawing, measurement and report stay linked. Our guides on inspection documentation best practices and inspection report formats cover what to keep and how to structure it, and the broader picture is in quality control in manufacturing.

How QA-Report supports the plan in daily use

QA-Report is a cloud-based inspection and reporting platform for manufacturers in aerospace, automotive and medical device production. Its measurement wizard links ballooned drawing dimensions to measured results and flags out-of-tolerance deviations automatically. CMM data import, batch and serial tracking and the document vault keep the evidence for each control plan row in one place, and its FAI, AS9102 and PPAP reports are built for audit use.

Plans start at $0 per month on Free, with Basic at $49.99 per month and PRO at $149.99 per month. On-Premise and Enterprise Solutions are available for organizations with confidentiality requirements, with pricing on request. See QA-Report pricing for what each plan includes, or look at CMM inspection software for automated reports if your measurements come from a coordinate measuring machine.

QA-Report

Perspective

When we talk with quality engineers about control plans, the problem is rarely the template. It is the gap between what the plan says and what the inspection records show. A beautifully formatted table with a frequency of one part per hour means little if the hourly result lives on a paper sheet in a drawer.

That is why we build around a connected chain: ballooned drawing, measured value, tolerance check, report. When every control plan row can point to a measured result that is already linked to its drawing dimension, auditors see traceability, not a promise. Teams that also manage parts, batches and operations in the same platform keep rejection and recovery records alongside the inspections that triggered them, which is exactly what a reaction plan needs.

Our advice is simple. Keep the plan short, risk-based and revision-controlled, and make sure the evidence behind each row is one click away.

Michael Chen

FAQ

What is the standard format for a control plan?

There is no single mandated layout, but most plans follow the AIAG form. A header with plan identification and phase, then rows with process step, machine, characteristics, specification, measurement method, sample size and frequency, control method and reaction plan. Aerospace plans add key characteristic or critical item classification.

How many control plans do I need for one part?

Up to three over the product life cycle: prototype, pre-launch and production. Each is a revision of the same logic with tighter or looser controls as the process is proven. Your customer may also accept a generic plan covering a product family instead of one per part number.

What is the difference between a control plan and a PFMEA?

The PFMEA identifies and ranks what can go wrong in each process step. The control plan states how you will detect or prevent it and what you do when a check fails. The PFMEA ranking drives how much control each row needs.

How often should a control plan be updated?

Treat it as a live document. Update it after any product or process change, after shipping nonconforming product, after customer complaints with corrective action, and on a risk-based schedule. Confirm any customer-specific timing in your customer's requirements.

Sources