A molded part is not well controlled because many dimensions are measured; it is controlled when the checks answer the questions that matter to its intended use. A part-specific inspection plan connects the current drawing and approved samples to the characteristics, methods, timing and reaction needed to make a release decision. For OEM, procurement and product teams, it also sets a common basis for reviewing an injection molding quality report rather than comparing isolated results. The plan should be agreed for the individual part and its application, not copied wholesale from another programme.
Measurement capability belongs in the plan: the instrument, fixturing, datum strategy and decision rule can affect whether a result is trustworthy.[3]
01
Start with function and turn it into a controlled characteristic set
Begin with the released part definition: drawing revision, CAD where relevant, material and colour direction, appearance standard, mating-part information and any approved sample or boundary example. Then ask what could prevent the part from assembling, sealing, locating, carrying load, operating safely, presenting the intended appearance or being traceable in the buyer’s process. A key dimension with no functional consequence may warrant periodic confirmation, while a small interface feature may need closer attention. ISO’s plastic-moulded-parts standard addresses manufacturing tolerances and acceptance conditions, while recognising that functional needs can require additional dimensional and geometrical specification.[1]
Record each agreed critical-to-quality characteristic in plain language alongside the drawing reference and acceptance criterion. This can include a datum-related feature, mass only where mass is meaningful, gate vestige location, colour comparison condition, insert presence, thread engagement, leak performance or packaging label. Do not label every note a CTQ: doing so dilutes attention and can make the plan expensive without improving the decision. The term injection molding CTQ should mean a characteristic whose variation could materially affect the stated part requirement. Its priority, method and frequency remain part-specific, especially when the part’s function or end-market changes.
02
Choose a measurement method that can support the acceptance decision
For every characteristic, name the method, instrument or gauge, fixture condition, measurement location and reference datum—not merely ‘inspect’. A caliper can be useful for a broad external check, but it may be unsuitable for a deep internal feature, a profile requirement or a tight positional relationship. Visual criteria need the same discipline: define viewing condition, distance or reference standard where those details matter to the application. NIST notes that measurement equipment is one principal source of dimensional uncertainty and that equipment-selection rules should be based on design information and measurement requirements.[3]
A plan should also say who records the result and how the method is verified or maintained. This does not require one universal device or numerical gauge rule for every molding project. It requires a defensible match between the feature, tolerance, access, material condition and decision being made. Where a feature is susceptible to part temperature, conditioning, fixturing force or operator interpretation, capture the agreed condition in the work instruction. That prevents a later dispute in which two parties use different set-ups and obtain different values from the same part.
03
Set inspection timing and sampling by risk, not by habit
Separate launch evidence from routine control. At launch, a first-off or first-article review can confirm that the actual tool, material and agreed measurement set-up are being assessed against the released definition. In some supply chains this is requested as an injection molding FAI; the required content, sample status and approval route should be stated in the purchase or quality agreement rather than assumed. A later routine plan may use start-up checks, in-process checks, final checks or lot-based sampling. These activities serve different decisions, so one record should not automatically replace another.
Sampling frequency should follow the consequence of escape, the stability and visibility of the characteristic, the ability to detect it later, and the information available from process history. ISO 2859-1 describes acceptance-sampling systems for inspection by attributes; it is a framework, not a substitute for choosing the right characteristic or a reliable method.[2] FDA guidance written for drug manufacturing offers a useful general planning discipline: specify sampling points, number of samples and frequency by operation and attribute, with confidence selected through risk analysis.[5] It is not a general plastics requirement. For a molded component, the buyer and supplier should agree the approach that fits the part and contractual context.
| Inspection decision | Typical evidence | Planning question for the buyer |
|---|---|---|
| Initial sample or first article | Dimensional and visual results tied to the released revision; material and process context where agreed. | What must be approved before routine production, and which deviations may be accepted in writing? |
| Start-up or changeover confirmation | Checks after a defined set-up, material, tool, parameter or personnel event. | Which events require re-confirmation, and who may authorise restart? |
| In-process control | Recorded checks on selected characteristics during the run, with a stated frequency and reaction plan. | Which characteristics can drift or become hidden before final inspection? |
| Final or lot release | A defined sample or verification of completed product, packaging and identification. | What is the lot definition, and what evidence travels with the shipment? |
| Periodic capability review | Trend data or focused study of a significant characteristic when the project warrants it. | Which evidence would justify changing frequency, method or containment? |
04
Link the plan to process knowledge, change control and response
An inspection plan is stronger when it sits beside a simple process map: incoming resin or inserts, drying or preparation where applicable, molding, any trimming or assembly, inspection, packing and release. At each step, identify the condition that could create the characteristic being checked and the response if a result is outside the agreed criterion. AIAG presents control plans alongside FMEA, measurement-system analysis, statistical process control and PPAP as connected quality-planning tools; the exact automotive formats are not mandatory for every project, but the linkage is useful.[4] It helps the team avoid treating inspection as the only control available.
Use injection molding process validation carefully. In regulated or customer-defined programmes, it may have a formal protocol and evidence expectation. In other programmes, the practical objective may simply be documented confidence that the defined process can repeatedly make parts against the agreed requirements. FDA frames validation as a lifecycle concept and distinguishes process qualification from continued process verification for its drug-manufacturing context.[5] That model can inform questions about documented conditions, data review and changes, but it does not by itself impose a validation obligation on an ordinary molded part. Define the requested evidence, scope and acceptance basis up front.
05
Make the report usable for release and escalation
Decide before production what a buyer needs to see: part number and revision, lot or run identity, sample size, characteristic results, units, method where material to interpretation, disposition, date and responsible reviewer. Include a clear reference to any approved deviation, rework instruction or containment action. A report that lists measurements without linking them to the released acceptance basis can be difficult to audit or use in a receiving decision.
The reaction plan is as important as the check. State what happens when a result is suspect or nonconforming: pause, segregate, expand review, notify, investigate, obtain disposition, re-inspect or resume under an agreed condition. Identify the trigger for updating the plan after a tool repair, material change, design revision, recurring issue or changed requirement. This turns the document into a living agreement between product, quality and procurement teams. It also preserves a more useful audit trail than a one-time checklist created only for launch.
Practical takeaway
Before you move
to the next review.
- Freeze the drawing revision, material direction, appearance references and mating-part assumptions that define acceptance.
- Identify the limited set of function-driven characteristics and write an unambiguous criterion for each one.
- Assign a method, datum or fixture condition, timing, record and accountable owner to every planned check.
- Agree launch, routine, change-triggered and final-release evidence separately, including the lot definition and sampling basis.
- Document the reaction path for suspect or nonconforming results, deviations, containment, communication and plan revision.
FAQ
Common questions
Does every drawing dimension need the same inspection frequency?
No. The frequency should reflect the feature’s function, risk, measurement practicality and the process knowledge available for that specific part. The agreed plan should make the rationale visible rather than implying that every characteristic receives identical treatment.
What should a buyer request in a first-article package?
Request only evidence that supports the release decision: the controlled part revision, agreed characteristics and results, method context where needed, sample or lot identity, and any deviations. Add material, appearance, assembly or process information when the project’s requirements make it relevant.
Can an AQL-style sampling plan prove every part is acceptable?
No sampling plan proves every individual unit is conforming. Attribute-sampling systems provide a structured lot-inspection approach, but they must be paired with suitable requirements, methods, lot definition and response rules. The right approach is part- and contract-specific.[2]
References
Source material
- [1] ISO, “ISO 20457: Plastics moulded parts — Tolerances and acceptance conditions.”
- [2] ISO, “ISO 2859-1: Sampling procedures for inspection by attributes.”
- [3] National Institute of Standards and Technology, “Rule Model for Selecting Dimensional Measurement Equipment in Inspection Planning.”
- [4] Automotive Industry Action Group, “Quality Core Tools.”
- [5] U.S. Food and Drug Administration, “Process Validation: General Principles and Practices.”
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.