Rules
Part of Graphic apparel production guide: brief, garment, artwork, method, proof, print, and records
Graphic apparel production problems: unclear briefs, wrong blanks and print drift
Graphic apparel production problems like unclear briefs, wrong blanks and print drift get traced to a cause, contained and stopped from repeating.
What to take away
- Freeze the press and the paperwork before anyone sorts, reprints or cleans anything.
- Find the last verified good garment and the first bad one. Everything between them is suspect.
- Sort the cause into four bucketsspec, material, process drift, record failure.
- A corrected print does not close the job while affected inventory is still loose.
- Keep the original brief, artwork, proof and traveler after you add corrected versions.
Name the order before you argue about the print
Two people can look at the same shirt and describe different defects. Settle identity first. This includes order code, artwork revision, blank style, and process. Also unit range, time, operator, settings, and who approved it. Include the physical reference garment too.
Order identity checklist
- Order code and artwork revision
- Blank style and process
- Unit range, time, operator
- Settings and approver
- Physical reference garment
Contain the run
Stop the process. Segregate unprinted, in-process, accepted, rework and rejected units. Mark the containers so nothing reaches fulfillment.
Unit timeline evidence
- Last known goodInspection record and garment
- First known badDefect record and garment
- Change betweenAdjustment, stop, material, operator
- Potential populationEvery unit since last check
- Immediate dispositionHold, inspect, rework, reject, release
Lead-in: Preserve these in the state the investigation needs:
- screens
- files
- transfers
- ink lots
- press settings
- test results
- physical samples
Build a unit timeline:
Point / Evidence
- Last known good
- Inspection record and the identified garment
- First known bad
- Defect record and the identified garment
- Change between them
- Adjustment, stop, material addition, cleaning, operator, environment
- Potential population
- Every unit since the last effective check
- Immediate disposition
- Hold, inspect, rework trial, reject, or qualified release
Do not ship apparently good units from the affected window until the risk and the inspection method are resolved.
Unclear brief
The symptom is disagreement about size, placement, color, softness, deadline or allowed variation. Compare the signed brief against messages, mockups, samples, the quote and the proof. Establish which document held authority when production started.
Vague brief vs measurable spec
Vague brief
- Size
- "Looks right"
- Placement
- "Centered"
- Color
- "Match sample"
- Softness
- "Soft hand"
- Variation
- "Close enough"
- Changes
- Unclear
Measurable spec
- Size
- Finished dimensions
- Placement
- Datum and tolerance
- Color
- Approval method
- Softness
- Surface criteria
- Variation
- Count and test method
- Changes
- Named approver
Swap adjectives for measurable or physical references. Use finished dimensions, datum, tolerance, and color approval method. Also use surface criteria, test method, count, and who approves a change. If customer and shop read an omitted field differently, do not invent a retroactive requirement.
Wrong blank
Lead-in:
Blank verification checklist
- Supplier and style code
- Color, size, fiber label
- Fabric structure and batch
- Measurements and seams
- Coating and care label
- Check supplier, style code, color, size, fiber label, fabric structure, batch, measurements, seams, coating and care label. A substitute can look close while responding differently to heat, ink, pretreatment, adhesive, washing or stretch.
Quarantine decorated and undecorated stock. Test whether the actual blank meets the approved product requirements without misdescription. If it cannot, replace it or get a documented redesign and customer approval. Fix purchasing and receiving controls, not just the traveler.
Late artwork change
Preserve the previously approved file. Identify which screens, transfers, proofs and garments used each version. A filename like "final-new-2" is not control.
Issue a numbered revision carrying a change description, owner, date, and effect on separations or method. Also include old-file disposition, new proof requirement, cost, schedule, and approval. Block obsolete assets physically and digitally.
Missed physical proof
A digital mockup leaves several details unresolved:
- physical scale
- garment color
- ink hand
- opacity
- underbase
- print texture
- heat marks
- placement behavior Find out whether the contract required a sample or first-off, and who released the run without it.
Five stages in conventional textile printing:
- design concept to artwork
- approval
- engraving or screen preparation
- color matching
- strike-off
- a second approval
- then production That academic textile-printing workflow comparison covers broader rotary and digital textile processes, so it supports approval gates rather than an exact T-shirt schedule.
Put a physical gate where remaining risk can still be corrected. Define who approves, what the proof establishes, and which changes invalidate it.
Print drift
Classify drift by direction and by time. Placement creeps, registration separates, opacity falls, color shifts. Detail closes, adhesion weakens. Plot the defect against unit order instead of mixing the stack.
Check what changed: garment surface, platen loading, platen adhesive, screen tension.
- clamps
- off-contact
- squeegee
- ink viscosity
- deposit
- flash
- temperature
- dwell
- pressure
- transfer condition
- operator
- environment Use measurements and identified samples.
Weak records
Reconstruct from the most direct evidence:
- purchase and receiving records
- source art
- approvals
- machine logs
- screen or transfer IDs
- timestamps
- inspection sheets
- photographs
- packing counts
- shipping records Label estimates and recollections as estimates and recollections.
ASQ describes root cause analysis as finding underlying causes so action prevents recurrence, and its beginner article lists data collection, causal-factor charting, cause identification, recommendation and implementation. Use that root-cause process summary without claiming a five-why discussion alone proves causation.
Add corrections as dated entries. Never overwrite the original record. Then test whether the new gate, identifier, measurement or training would have caught this failure before production.
Closure criteria
- affected population identified and dispositioned;
- customer and regulatory notifications completed where required;
- physical correction verified against the specification;
- root cause supported by evidence;
- corrective action assigned and implemented;
- effectiveness checked on later work;
- cost, schedule, inventory and records reconciled.
Common questions
Can good-looking shirts inside a held range be shipped?
Only after an approved inspection or test establishes conformity and traceability for those units. Appearance alone does not release them.
Should the newest artwork file replace every older version?
No. Preserve history, block obsolete versions from use, and record which units used each version.
Is operator error a sufficient root cause?
Usually not. Ask how these factors let the error reach production:
- instructions
- interface
- training
- workload
- verification
- equipment
- change control
When can production restart?
After the cause is controlled, materials and settings are verified, a new first-off passes, and authorized release is recorded.
