Build an automotive injection molding control plan by linking each meaningful product or process risk to a defined characteristic, control method, inspection frequency, reaction, responsible role and retained record. Start with the latest drawing, process flow and risk analysis, then make the plan specific enough that an operator, quality engineer and buyer would make the same decision when a result changes. AIAG describes APQP/control plan, FMEA, MSA, SPC and PPAP as connected quality core tools rather than isolated documents.[1]
For Europe, the United States and Middle East supply programmes, the customer’s drawing, purchase specification, customer-specific requirements and approval conditions remain the governing inputs. A control plan is evidence of planned process control; it is not, by itself, a certification or a substitute for product validation.
A control plan should answer five practical questions for every important characteristic: what can go wrong, how will we detect it, how often will we check, what happens when the result is unacceptable, and which record proves the check occurred?
01
Start with the product risk and the process flow
A control plan should be built from the actual process sequence: material receipt and drying where applicable, mold setup, first-off approval, molding, in-process checks, visual inspection, secondary operations, packing and release. Map each step to the potential failure modes identified in the process FMEA. If the plan lists only final dimensions, it may miss upstream causes such as resin identity, moisture condition, cavity balance, mold temperature, insert presence or an unapproved parameter change.
Use one controlled part and process revision across the drawing, PFMEA, process flow, work instructions, inspection forms and approval package. The AIAG core-tools framework explicitly connects control plans with FMEA, MSA, SPC and PPAP.[1] In practice, that means a changed risk rating, customer requirement or special characteristic should trigger a documented review of the related control-plan row rather than an informal spreadsheet edit.

02
Translate special characteristics into visible controls
Special characteristics are product or process features that warrant additional attention because their variation could affect safety, regulatory obligations, fit, function, performance or downstream manufacturing. The customer may define symbols and names, or may require the supplier to propose them for approval. Do not invent a universal symbol set. Carry the customer’s designation into the drawing review, PFMEA, control plan, work instruction and inspection record, and resolve any mismatch before production release.
For an injection-molded part, a special characteristic might be a sealing interface, connector datum, retention feature, wall area affecting a safety function, material identity or a process setting that directly influences a critical result. The plan should state the characteristic, specification or target, measurement system, control method, sampling rule, reaction and record. A characteristic can be important without automatically requiring 100% measurement; the decision should follow risk, detectability, capability evidence, customer requirements and the consequences of escape.
| Control-plan field | Decision to make | Evidence to retain |
|---|---|---|
| Characteristic and classification | What feature or process input is controlled, and which customer or internal designation applies? | Drawing revision, requirement source and approved symbol/key. |
| Specification or target | What limit, nominal, attribute rule or approved visual standard is used? | Controlled specification, master sample or acceptance standard. |
| Method and equipment | Which gauge, fixture, test or validated visual method will produce the result? | Gauge ID, method revision and applicable MSA evidence. |
| Frequency and sample | When is it checked, on how many pieces, and after which events? | Sampling instruction, lot/shift/cavity context and completed results. |
| Reaction and disposition | What is contained, who is notified, and who can release affected product? | Reaction record, traceability, disposition and corrective-action link. |
03
Use SPC where the data and process support it
Statistical process control is not simply a request to add a chart to every row. Select a variable or attribute chart when the characteristic, measurement system, subgrouping and data stream make the chart meaningful. NIST explains that control charts routinely monitor quality and distinguish common variation from signals that warrant investigation; a point outside a limit is not the only concern, because systematic non-random patterns can also indicate a problem.[2]
For a molded part, a chart may be appropriate for a dimensional characteristic measured by a capable method, or for a process variable that has a demonstrated relationship to output. Define the subgroup, sampling interval, chart type, baseline or limits, and the person who reviews the signal. Keep specification limits separate from control limits: specification limits describe product acceptance, while control limits describe observed process behavior. ISO 7870-2 provides guidance on the Shewhart approach, while its ISO page notes that the 2013 edition is withdrawn and identifies ISO 7870-2:2023 as the newer version.[3]
Do not claim capability from a short, unstable or poorly measured run. Confirm the measurement method and its suitability before interpreting SPC results. If the process is not yet stable, the plan should say how data will be collected and what interim containment applies instead of presenting a misleading capability number.

04
Make the reaction plan immediate, specific and traceable
A reaction plan is the action path after a failed inspection, SPC signal, machine alarm, material discrepancy, visual defect or customer complaint. It should be written in operational language: stop or hold the process when appropriate, identify the last known good check, segregate potentially affected parts, verify the measurement or condition, notify the named role, investigate the cause and obtain authorized disposition before release. The correct action depends on the risk and customer agreement; not every signal requires the same containment scope.
Name the boundary of affected product. A useful record can connect part number and revision, mold and cavity where relevant, machine, material lot, time window, operator or shift, inspection result, quantity held, decision maker and disposition. For a safety- or regulatory-relevant characteristic, escalation should follow the customer’s reporting and approval rules. A supplier should not promise unilateral concession authority when the purchase agreement assigns that decision to the OEM or tier customer.
05
Set inspection frequency from risk, evidence and change points
Inspection frequency should be justified rather than copied from a generic template. Consider characteristic risk, process maturity, demonstrated stability and capability, measurement burden, cavity-to-cavity behavior, lot size, shift changes, tool maintenance, material changes, setup approval and customer-specific requirements. Typical control-plan events may include first piece or first-off approval, start-up after a changeover, periodic in-process checks, end-of-run verification and layered or audit checks. These are planning categories, not universal frequencies.
Increase control after a new tool, mold repair, engineering change, material change, prolonged stoppage, abnormal trend, supplier change or complaint until objective evidence supports the agreed normal plan. Records should be legible, attributable and retrievable for the period required by the customer, applicable quality system and contract. IATF’s official site provides access to customer-specific requirements and related automotive oversight resources, reinforcing that a generic control plan cannot replace the applicable customer rule set.[4]
For procurement review, ask the supplier to show a sample completed record rather than only a blank form. The sample should make it possible to trace the result to the part revision, lot or time window, equipment or gauge, person performing the check, acceptance decision and reaction if the result was not acceptable.

Practical takeaway
Before you move
to the next review.
- The control plan is linked to the current drawing, process flow and PFMEA revision.
- Customer-designated special characteristics and internal risk features are identified with an agreed symbol or key.
- Each critical row names the method, gauge or test, sample rule, frequency, responsible role and record.
- SPC use distinguishes specification limits from control limits and defines what happens after a signal.
- Reaction records can identify affected product, containment, disposition, approvals and follow-up action.
FAQ
Common questions
Does every injection-molded dimension need SPC?
No. SPC should be selected where the characteristic, measurement system, data structure and process behavior support useful statistical monitoring. Other controls may be more appropriate for attributes, low-volume features, validated tests or characteristics governed by a customer-specific rule.
Should a special characteristic always be inspected 100%?
Not automatically. The decision should consider risk, detectability, capability evidence, process controls, customer requirements and the consequences of escape. If sampling is proposed, document the rationale and approval rather than assuming that a generic percentage is acceptable.
What should happen when an SPC chart signals but the part still meets drawing limits?
Follow the defined reaction plan. Verify the measurement and charting conditions, evaluate the signal and process cause, and determine whether containment or increased monitoring is needed. A part within specification is not proof that the process is behaving predictably.
Who approves the reaction plan?
The supplier should define the operational response with the product owner or customer quality contact. Release, deviation, concession and safety-related escalation authority must follow the contract, customer-specific requirements and applicable approval process.
How long should control-plan records be kept?
Use the longest applicable period required by the customer, purchase terms, approved quality agreement and the supplier’s governing quality system. State the retention rule in the record-control procedure and make retrieval practical for audits, complaints and traceability reviews.
References
Source material
- [1] AIAG, Quality Core Tools: APQP, Control Plan, PPAP, FMEA, SPC and MSA
- [2] National Institute of Standards and Technology, NIST/SEMATECH e-Handbook: Process or Product Monitoring and Control
- [3] International Organization for Standardization, ISO 7870-2:2013 Control charts — Part 2: Shewhart control charts; page identifies ISO 7870-2:2023 as the newer version
- [4] International Automotive Task Force, IATF 16949 and customer-specific requirements resources
This guide is general information for early procurement and engineering discussion. It does not replace product-specific DFM, qualification, validation, regulatory assessment or legal advice.