Control Plan Template: Linking PFMEA to Production Controls
The control plan is where your PFMEA meets the production floor. Every detection control you identified in the PFMEA should have a corresponding control plan entry that specifies the method, frequency, sample size, and reaction plan. This guide provides the complete control plan template structure, explains how each column connects to your PFMEA, and shows worked examples for automotive manufacturing.
For PFMEA fundamentals, see our PFMEA manufacturing example. For IATF 16949 requirements governing control plans, see our IATF 16949 FMEA Requirements guide.
What Is a Control Plan?
A control plan is a document that describes the system of controls for a manufacturing process. It specifies what to check, how to check it, how often, and what to do when a check fails. In the APQP framework, the control plan is one of the five core tools (alongside FMEA, SPC, MSA, and PPAP).
The control plan is:
- A required PPAP deliverable for automotive OEMs
- The operational translation of your PFMEA risk analysis
- The basis for work instructions, inspection plans, and SPC charting
- A living document updated when the process, product, or FMEA changes
The chain is: Process Flow Diagram → PFMEA → Control Plan → Work Instructions. Every link must be traceable.
Three Phases: Prototype, Pre-Launch, Production
| Phase | Purpose | Controls |
|---|---|---|
| Prototype | Validate design during prototype build | Dimensional measurements, material testing, and performance tests. Higher sampling (often 100%). Focus on design verification. |
| Pre-Launch | Validate process during trial production runs | All production controls active plus enhanced inspection. SPC studies to confirm process capability. Machine and gage R&R studies. Validates Cpk targets. |
| Production | Ongoing production control | Reduced frequency based on demonstrated process capability. Statistical sampling per SPC rules. Poka-yoke and automation where validated. Reaction plans for out-of-control conditions. |
Most audit findings involve the transition: pre-launch controls that were supposed to carry into production but were relaxed without process capability data to justify the reduction.
Control Plan Template: Column-by-Column
| # | Column | Purpose | Source |
|---|---|---|---|
| 1 | Control Plan Number | Unique document identifier, links to PPAP | QMS document control |
| 2 | Part Number / Rev Level | Which product revision this applies to | Engineering drawing |
| 3 | Part Name / Description | Product identification | Bill of materials |
| 4 | Supplier / Plant | Manufacturing location | PPAP submission |
| 5 | Control Plan Phase | Prototype / Pre-Launch / Production | APQP timing |
| 6 | Process Step / Op # | Matches the process flow diagram and PFMEA operation | PFD / PFMEA |
| 7 | Process Name / Description | Operation description | PFD |
| 8 | Machine / Device / Fixture | Equipment used at this station | PFD / PFMEA |
| 9 | Product Characteristic | What product feature is being controlled (dimension, material property, appearance) | Drawing / DFMEA |
| 10 | Process Characteristic | What process parameter controls the product characteristic (tonnage, temperature, speed) | PFMEA |
| 11 | Special Characteristic Symbol | OEM-specific symbol (Ford diamond, GM shield, etc.) | DFMEA / OEM CSR |
| 12 | Specification / Tolerance | Engineering tolerance or accept/reject criteria | Drawing |
| 13 | Evaluation / Measurement Method | What tool or technique is used to check (CMM, go/no-go gage, SPC chart, visual) | MSA / PFMEA detection control |
| 14 | Sample Size | How many pieces are checked (1 pc, 5 pc, 100%) | SPC plan / capability |
| 15 | Sample Frequency | How often (every hour, every setup, per coil, per lot) | SPC plan / capability |
| 16 | Control Method | How results are recorded and monitored (X-bar R chart, p chart, checklist, automated sensor) | SPC plan |
| 17 | Reaction Plan | What to do when the control fails or the characteristic is out of specification | PFMEA / QMS procedure |
| 18 | PFMEA Reference | PFMEA row or failure mode that drove this control | PFMEA |
Header Information
In addition to the row-level columns, the control plan header must include:
- Control plan number and revision date
- Customer / OEM name
- Part number, name, and revision level
- Supplier name, code, and plant location
- Core team members
- Date (original) and date (revision)
- Customer engineering approval (if required)
How the PFMEA Drives the Control Plan
The connection is direct: every detection control in the PFMEA must appear in the control plan with concrete operational details.
| PFMEA Column | → | Control Plan Column |
|---|---|---|
| Process Step / Op # | → | Process Step / Op # (same number) |
| Failure Mode | → | Context for the characteristic being checked |
| Failure Effect (Severity) | → | Determines whether special characteristic symbol applies |
| Detection Control | → | Evaluation / Measurement Method |
| Detection Score | → | Informs sample size and frequency (lower D score = higher capability) |
| Action Priority | → | Drives level of control: High AP = enhanced control. Low AP = standard monitoring |
If the PFMEA says “CMM check of hole true position” as a detection control with D=3 (high capability), the control plan entry should show: method = CMM, sample size = 5 pc, frequency = per hour, control method = X-bar R chart with Cpk ≥ 1.67 target. The PFMEA identifies what to detect, the control plan specifies how, how often, and what happens when it fails.
Worked Example: Stamping Operation
Using the B-pillar bracket from our PFMEA example:
| Op | Product Char. | Spec | Method | Sample | Freq | Control | Reaction Plan |
|---|---|---|---|---|---|---|---|
| 10 | Material grade | HSLA 590 | Spectrometer + cert review | Per coil | Every coil | Incoming log | Quarantine coil, contact supplier, do not process until verified |
| 10 | Coil thickness | 1.8 mm ± 0.05 | Micrometer at 3 points | Per coil | Every coil | Incoming log | Reject coil if out, escalate if 2 consecutive coils fail |
| 20 | Blank dimensions | 420 × 180 ± 0.5 | Go/no-go fixture | 5 pc | Per setup + hourly | SPC X-bar R | Stop, adjust blanking die, re-verify 5 consecutive OK |
| 30 | Bend angle | 90° ± 0.5° | Digital protractor | 5 pc | Per setup + hourly | SPC X-bar R | Stop, adjust tonnage, re-verify Cpk ≥ 1.33 |
| 30 | Crack at bend | No visible cracks | Eddy-current inline | 100% | Continuous | Auto-reject bin | 3 consecutive rejects: stop press, investigate die and material |
| 40 | Hole true position | ± 0.2 mm TP | CMM | 5 pc | Hourly | SPC X-bar R | Stop, inspect locator pins, 100% sort since last good check |
| 50 | Weld strength | ≥ 15 kN push-out | Weld monitor + destructive test | 1 pc destruct | Every 200 pc | Weld monitor log + destruct log | Quarantine from last good test, 100% sort, electrode dress check |
| 50 | Nut perpendicularity | ± 1.5° | Go/no-go fixture | 5 pc | Per setup + hourly | Checklist | Stop, adjust fixture, re-verify 5 consecutive |
| 60 | E-coat thickness | 20-30 μm | Magnetic gauge | Per batch cert + 3 pc spot check | Per incoming batch | Incoming cert log | Reject batch, escalate to e-coat supplier |
| 70 | Final dimensional audit | Per drawing GD&T | CMM full layout | 1 pc | Per shift | Layout inspection report | Hold production, investigate and sort if any dimension fails |
Reaction Plans: The Most Neglected Column
The reaction plan is the most commonly under-specified column in control plans. “Inspect and adjust” is not a reaction plan. A complete reaction plan specifies:
- Containment: What happens to suspect product? (Quarantine, tag, segregate since last known good check)
- Disposition: How is suspect product evaluated? (100% sort, sample re-inspection, functional test)
- Root cause: What investigation method is triggered? (5-Why, 8D, Ishikawa)
- Restart criteria: What must be true before production resumes? (N consecutive OK parts, re-validated Cpk)
- Notification: Who is informed? (Shift lead, quality engineer, customer if escaped)
Auditors often pick a control plan entry, ask the operator to describe the reaction plan, and compare the answer to the documented plan. A mismatch is a finding.
Special Characteristics in the Control Plan
Special characteristics - identified in the DFMEA and carried through the PFMEA - must be explicitly marked in the control plan. The marking convention depends on the OEM:
| OEM | Symbol | Meaning |
|---|---|---|
| Ford | ◊ (diamond), ◊◊ (double diamond) | Significant / Critical characteristic |
| GM | Shield, inverted delta | Safety / Fit-Function characteristic |
| Stellantis | Pentagon, shield | Operator safety / Product safety characteristic |
| VW Group | D, TLD, Safety | Documentation / Safety characteristic |
For special characteristics:
- Enhanced controls are required (tighter sampling, error-proofing, SPC with demonstrated Cpk ≥ 1.67)
- The characteristic must flow through all linked documents: DFMEA → PFMEA → Control Plan → Work Instruction → Inspection Plan
- Records must demonstrate ongoing process control (SPC charts maintained)
Common Control Plan Mistakes
“100% visual inspection” as the only control for a critical dimension. Visual inspection cannot reliably detect a 0.2 mm hole position deviation. Match the control method to the characteristic tolerance and measurement system capability.
Frequency not justified by process capability. If your process Cpk is 1.05, sampling 5 pc per shift is insufficient. The sampling frequency must be justified by demonstrated capability. If Cpk is marginal, sampling must be tighter.
Missing operations. The control plan must cover every operation on the PFD, including rework, repair, and outsourced operations. E-coat at a subcontractor still needs a control plan entry (even if it is incoming inspection of the returned batch).
Generic reaction plans across all characteristics. “Stop and notify quality” is a default, not a plan. Safety-critical characteristics require immediate containment and lot segregation. Non-critical characteristics may need only SPC investigation. Differentiate.
No revision after PFMEA update. If the PFMEA was revised - new failure mode, new control, or changed action priority - the control plan must be revised to match. This is the most common IATF 16949 audit finding for control plans.
What Auditors Check
- Every PFD operation has a corresponding control plan entry
- Every PFMEA detection control maps to a specific control plan row
- Special characteristics are marked with correct OEM symbols
- Measurement methods match the MSA (gage R&R results support the measurement capability)
- Sample sizes and frequencies are justified by process capability data
- Reaction plans are specific, not generic
- Operators can describe the reaction plan when asked
- Revision history shows updates concurrent with PFMEA revisions
- The control plan is the current version posted at the workstation
How Tacit AI Approaches This
Tacit AI generates control plan entries directly from your PFMEA detection controls, keeping the triad (PFD → PFMEA → Control Plan) synchronized.
PFMEA drives control plan generation. When you define a detection control in the PFMEA, the corresponding control plan entry is drafted with method, sample size, frequency, and reaction plan fields pre-populated from your process data.
Changes propagate. Update a PFMEA row - new failure mode, revised detection control, or changed Action Priority - and the control plan flags the affected entries for review. No manual cross-referencing.
Reaction plans from work order history. The reaction plan suggestions draw from your actual production data: what containment actions worked, what root causes were found, what restart criteria your team used. Not generic text, but your organization’s operational knowledge.
OEM format export. Control plans export in Ford, GM, Stellantis, and VW Group formats, with correct special characteristic symbols and required header information.
Building a control plan from your PFMEA? Book a working session and see Tacit AI generate the linked triad from your process flow. For the PFMEA input, see our PFMEA manufacturing example.
Frequently Asked Questions
Do I need a separate control plan for each product?
Yes, unless products share the exact same process, tooling, and characteristics. Family control plans are acceptable when products are truly similar (same part family, same equipment, same critical characteristics), but each product must be explicitly listed and any product-specific characteristics must have their own entries.
How often should the control plan be updated?
Every time the PFMEA is updated, the process changes, a customer complaint triggers a new control, or during scheduled reviews (at least annually). IATF 16949 Clause 8.5.1.1 requires the control plan to be reviewed and updated after any process or product change.
Can I reduce inspection frequency over time?
Yes, when supported by process capability data. If Cpk ≥ 1.67 is demonstrated and sustained over a statistically significant period, frequency reduction can be justified. The justification must be documented and the PFMEA detection score updated to reflect the reduced sampling.