A paper cup defect reference sample helps inspectors apply visual criteria consistently, but a physical “golden sample” can fade, deform, collect dust or lose its approval history. Reference control must define the defect, severity boundary and current authorized version.
This article supports visual-inspection control. It does not establish universal cosmetic limits or replace customer-approved specifications. Quality personnel must approve every reference.
Diagnostic decision table
| Reference issue | Control | Risk |
|---|---|---|
| Defect changes with storage | Condition and periodic review | Reference drifts silently |
| Only one extreme sample exists | Boundary examples and images | Minor defects are over-rejected |
| Identity is unclear | Unique ID and revision | Obsolete sample remains in use |
| Lighting varies | Defined viewing conditions | Appearance changes by workstation |
Define the defect and decision boundary
Write the defect name, location, appearance and severity rule using the approved specification. Separate different mechanisms such as print scuff, seam contamination, rim split and base deformation. One attractive cup is not a complete acceptance standard.
Select pass-boundary and fail-boundary examples where practical. Record the product construction and whether the reference applies across colours, artwork and sizes. Avoid using a severe defect alone, because inspectors may not learn how to judge borderline production.
Create controlled physical and digital references
Assign an ID, revision, approval, date and scope. Photograph the reference with controlled scale, orientation and lighting. Physical samples should be protected from compression, moisture, heat, light and handling damage.
Digital images improve access but can vary by camera, screen and colour settings. Use them as approved aids under stated conditions, not as automatic replacements for physical comparison. Preserve original files and editing history; do not enhance a defect until its severity changes.
Standardize inspection conditions
Define lighting, background, viewing distance, inspection time and permitted handling. Identify features that require magnification or measurement rather than unaided visual judgement. Train inspectors with randomized examples instead of showing the label first.
Confirm agreement across shifts and sites using an attribute agreement study. A reference set that only its creator can interpret is not effective. Record false accepts and false rejects separately because they lead to different product and cost risks.
Review stability and remove obsolete samples
Schedule review based on material stability, use and risk. Compare physical samples with approved images and records. Replace faded, crushed, contaminated or ambiguous examples through the same approval process.
Retrieve obsolete revisions from workstations and digital folders. Mark retired samples so they cannot return to use. When customer criteria change, map old and new boundaries and retrain before production release rather than quietly relabelling existing samples.
Connect reference decisions to process evidence
Record which reference and revision supported each inspection result. For recurring defects, preserve actual production samples and trace them to machine position, material and time. Do not use a reference sample to declare root cause.
Send HANNAI clear images, orientation, severity decision and process identity when equipment review is needed. Controlled references improve communication while the approved specification remains the authority for acceptance.
Related equipment and next checks
Review the paper cup production line and the related guide to visual defect inspection agreement study. The Engineering Notes archive connects these checks with wider machine planning.
Frequently asked questions
Is one perfect cup a sufficient golden sample?
No. Include controlled defect definitions and boundary examples.
Can photographs replace physical samples automatically?
Only when the approved method controls image and viewing limitations.
When should samples be replaced?
When stability, damage, revision or review evidence requires replacement.
Do references prove defect cause?
No. They support classification; root cause needs process evidence.
Review the evidence with HANNAI
Send your machine model, component identification and the observations described in this guide. Include the drawing or material reference and any before-and-after measurements so the required next check can be identified.