| name | Controlplan |
| description | Generate AIAG-format Control Plans linked to PFMEAs with proper inspection methods, frequencies, and reaction plans. Supports Prototype/Pre-launch/Production phases. USE WHEN user says 'control plan', 'CP', 'inspection plan', 'process control', 'sampling frequency', 'reaction plan', or 'special characteristics'. Integrates with PFMEA, AutomotiveManufacturing, and MSA skills. |
| effort | high |
Control Plan
When to Activate This Skill
- "Create a control plan for [part/process]"
- "What inspection frequency for [characteristic]?"
- "Define control methods for [operation]"
- "Link PFMEA to control plan"
- "What's the reaction plan for [out-of-spec condition]?"
- "Develop prototype/pre-launch/production control plan"
Control Plan Phases
Prototype Control Plan
- Purpose: Control during prototype build
- Focus: Dimensional measurements, material tests, performance
- Frequency: Often 100% on critical dimensions
- Duration: Until prototype approval
Pre-Launch Control Plan
- Purpose: Control during PPAP/validation runs
- Focus: Enhanced inspection, process validation
- Frequency: Increased sampling, SPC establishment
- Duration: Until production approval (PSW)
Production Control Plan
- Purpose: Ongoing production control
- Focus: Standardized controls, sustained quality
- Frequency: Based on process capability
- Duration: Life of production
AIAG Control Plan Format
Required Columns
| Column | Description | Source |
|---|
| Part/Process Number | Operation step number | Process Flow |
| Process Name/Description | Operation description | Process Flow |
| Machine/Device/Jig/Tool | Equipment used | Process documentation |
| Characteristic Number | Unique ID for characteristic | Sequential |
| Product Characteristic | Part feature being controlled | Drawing, spec |
| Process Characteristic | Process parameter affecting product | PFMEA |
| Special Char Class | CC, SC, or blank | PFMEA/Drawing |
| Product/Process Spec | Tolerance or requirement | Drawing, spec |
| Evaluation/Measurement | Gage/method used | MSA |
| Sample Size | Quantity inspected | Based on risk |
| Sample Frequency | How often inspection occurs | Based on risk |
| Control Method | How results are recorded/used | SPC, checklist, etc. |
| Reaction Plan | Action when out of spec | Defined response |
Special Characteristics
Critical Characteristics (CC)
- Symbol: Shield, (CC), or inverted delta
- Criteria: Safety or regulatory compliance
- Requirements:
- Documented in Control Plan
- Enhanced inspection (often 100%)
- SPC required (Cpk ≥1.67)
- Traceable records
- Identified on drawing
Significant Characteristics (SC)
- Symbol: Diamond, (SC)
- Criteria: Fit, function, appearance significant to customer
- Requirements:
- Documented in Control Plan
- Appropriate inspection frequency
- SPC recommended (Cpk ≥1.33)
- Records maintained
Standard Characteristics
- Symbol: None
- Criteria: All other characteristics
- Requirements:
- Documented in Control Plan
- Sampling based on risk
- Records as required
Sample Size and Frequency Guidelines
Based on Severity (from PFMEA)
| Severity | Sample Size | Frequency | Rationale |
|---|
| 9-10 (Safety) | 100% | Every piece | Safety critical - no escape |
| 7-8 (High) | 100% or n≥5 | Per hour or per lot | High customer impact |
| 5-6 (Moderate) | n≥3 | Per shift or per setup | Moderate risk |
| 1-4 (Low) | n≥1 | Per lot or periodic | Low risk |
Based on Process Capability
| Cpk Range | Typical Frequency | Notes |
|---|
| <1.00 | 100% or very frequent | Process not capable |
| 1.00-1.33 | Every hour or more | Marginal capability |
| 1.33-1.67 | Per shift or setup | Capable |
| >1.67 | Per setup, daily, or lot | Highly capable |
Standard Frequency Terms
| Term | Meaning | When to Use |
|---|
| Continuous | 100% inspection | Safety, high risk |
| Per piece | Every part | Critical characteristics |
| Per hour | n pieces every hour | SPC, high-risk |
| Per shift | Start and end of shift | Setup-sensitive |
| Per setup | At setup only | Stable characteristics |
| Per lot | One sample per production lot | Low-risk, stable |
| Daily | Once per day | Periodic verification |
| Weekly | Once per week | Very stable |
Control Methods
Inspection Methods
| Method | Description | Typical Use |
|---|
| Visual | Operator observation | Surface defects, presence |
| Go/No-Go Gage | Attribute check | Hole size, thread, profile |
| Caliper/Micrometer | Variable measurement | Dimensions |
| CMM | Coordinate measurement | Complex features |
| Height Gage | Surface plate measurement | Flatness, position |
| Profilometer | Surface finish | Ra, Rz |
| Hardness Tester | Material property | HRC, HRB |
| Torque Wrench | Fastener torque | Assembly |
| Leak Test | Seal integrity | Fluid systems |
| Functional Test | Performance verification | End-of-line |
Recording Methods
| Method | Description | When to Use |
|---|
| Check Sheet | Attribute recording | Visual checks, presence |
| Data Log | Variable data recording | Measurements |
| SPC Chart | Statistical control chart | Critical/significant |
| Traveler | Part-specific record | Traceability |
| Electronic | MES/database entry | High volume |
| None (process control) | Poka-yoke, automated | Error-proofed |
Reaction Plans
Standard Reaction Plan Format
IF [out-of-spec condition] THEN:
1. STOP - Halt production/segregate suspect
2. CONTAIN - Quarantine affected parts
3. NOTIFY - Alert supervisor/quality
4. EVALUATE - Determine root cause and extent
5. CORRECT - Implement fix before restart
6. VERIFY - Confirm correction effective
Reaction Plan Examples
Dimension Out of Spec:
- Stop production
- Segregate last good part to current
- Notify Quality/Supervisor
- Check tool condition, offset, fixture
- Adjust/replace as needed
- Run first piece, verify to spec
- Inspect segregated parts 100%
Visual Defect:
- Reject defective part
- Assess severity (isolated vs trend)
- If trend: notify supervisor, review process
- Identify root cause (tool, material, handling)
- Implement correction
- Resume with enhanced attention
SPC Out of Control:
- Mark control chart
- Continue production if within spec
- Investigate pattern (rule violated)
- Identify assignable cause
- Implement correction
- Document on chart
PFMEA to Control Plan Linkage
Required Flow
PFMEA Item → Control Plan Entry
───────────────── ───────────────────
Process Step → Part/Process Number
Function → Characteristic
Failure Mode → What we're preventing
Detection Control → Control Method
S/O/D Ratings → Sample Size/Frequency
Special Char → Special Char Class
Linkage Verification
For each Control Plan line, verify:
Output Format
When generating Control Plan content:
# Control Plan
**Part Number**: [P/N]
**Part Name**: [Description]
**Customer**: [Customer Name]
**Control Plan Number**: CP-[DEPT]-[SEQ]
**Phase**: [ ] Prototype [ ] Pre-Launch [X] Production
**Revision**: [Rev] | **Date**: [YYYY-MM-DD]
**PFMEA Reference**: PFMEA-[XXX]
## Operation: [Number] - [Name]
### Characteristic 1: [Description]
| Field | Value |
|-------|-------|
| **Char No.** | [Unique ID] |
| **Product/Process** | Product / Process |
| **Class** | CC / SC / - |
| **Specification** | [Nominal ± Tolerance] |
| **Evaluation Method** | [Gage/Equipment] |
| **Sample Size** | [n] |
| **Sample Frequency** | [When] |
| **Control Method** | [Recording method] |
| **Reaction Plan** | [Reference or inline] |
Integration with Related Skills
PFMEA
Control Plan is direct output of PFMEA:
- Every PFMEA detection control → Control Plan method
- PFMEA severity drives inspection frequency
- PFMEA updates trigger Control Plan review
Load: read ~/.claude/skills/Pfmea/SKILL.md
MSA
All gages in Control Plan require MSA:
- Gage R&R for variable gages
- Attribute agreement for attribute gages
- MSA acceptance before production approval
Load: read ~/.claude/skills/Msa/SKILL.md
AutomotiveManufacturing
Work instructions must reflect Control Plan:
- Inspection steps in work instructions
- Operator-level control methods
- Quality checkpoints documented
Load: read ~/.claude/skills/Automotivemanufacturing/SKILL.md
Workflow Routing
| Trigger | What to do |
|---|
| "create a control plan for [part / process]" | Pull PFMEA → enumerate ops → classify CCs/SCs → fill all 13 AIAG columns → emit phase-tagged CP markdown using the Output Format template above |
| "what inspection frequency / sample size for [characteristic]?" | Cross-check PFMEA Severity table + Cpk table; recommend the stricter of the two; cite both rules in the answer |
| "what's the reaction plan for [out-of-spec condition]?" | Match scenario to the Reaction Plan templates (Dimension OOS / Visual / SPC OOC); name responsibilities and timing explicitly — never "contact supervisor" alone |
| "link PFMEA to control plan" / "verify PFMEA-CP linkage" | Run the Linkage Verification checklist for every CP line; flag any missing failure mode, mismatched CC/SC class, or missing reaction plan |
| "review / audit my control plan" | Apply the Audit Preparation checklist from CLAUDE.md (current rev at point of use, MSA status, reaction-plan evidence, etc.) and report findings as a deficiency list |
| "prototype / pre-launch / production control plan" | Use the matching phase row in Control Plan Phases; default sampling and SPC commitments differ between phases |
| "special characteristic [CC / SC] — what controls?" | Apply the enhanced control requirements from CLAUDE.md (Cpk targets, MSA acceptance, error-proofing, traceability, annual review) |
Examples
Example 1 — New production control plan from PFMEA
User: "Create a production control plan for the damper rod machining operation. PFMEA shows OD diameter as a CC with Severity 9."
Output: Production-phase control plan markdown with one operation block — 100% CMM measurement (S=9 → no escape), Cpk ≥1.67 SPC requirement, X-bar/R chart with n=5 subgroups every hour, traceable record via MES, and a Type 1 "Stop and Contain" reaction plan referencing RP-CC-001 (segregate to last good, 100% sort suspect material, first-piece approval before restart).
Example 2 — Sampling frequency from severity + Cpk
User: "What sample size and frequency for a process characteristic with PFMEA Severity 7 and Cpk 1.45?"
Output: Stricter of the two rules wins. Severity 7-8 row says "100% or n≥5 per hour or per lot"; Cpk 1.33–1.67 says "per shift or setup". Recommendation: n≥5, per hour, X-bar/R SPC. Rationale citing both tables and the rule that high-severity overrides marginal capability advantages.
Example 3 — Reaction plan for an SPC trend, dimension still in spec
User: "My X-bar chart shows 7 points trending up but everything's still inside tolerance. What's the reaction plan?"
Output: SPC RP-SPC-002. Mark the chart (circle the rule violation, note WECO/Nelson rule number). Continue production since within spec. Investigate assignable cause — tool wear, fixture drift, incoming material. Implement correction. Document cause and action on the chart. Verify next 5 points return to control. Do NOT trigger Type 1 stop-and-contain because the part is still conforming.
Supplementary Resources
For detailed guidance:
read ~/.claude/skills/Controlplan/CLAUDE.md
For templates:
read ~/.claude/skills/Controlplan/templates/control-plan-template.md
For control method selection:
read ~/.claude/skills/Controlplan/reference/control-methods.md
For sampling guidelines:
read ~/.claude/skills/Controlplan/reference/sample-frequencies.md
