Tech Hub — Quality Control Standards

AQL Inspection Standards for Scarves — Sampling Plans, Defect Classification, and Pre-Shipment Inspection

Technical reference covering AQL definition, ISO 2859-1 sampling plan tables, AQL 2.5 lot-size-to-sample-size calculations, defect classification for scarves with specific examples, and pre-shipment inspection protocol requirements.

Data verified as of April 2026 — ISO 2859-1:1999, ANSI/ASQ Z1.4, industry inspection practice

AQL 2.5Most Common Apparel Standard

ISO 2859-1Primary Reference Standard

3 LevelsCritical / Major / Minor

GIL IIStandard Inspection Level

200+ pcsThreshold for AQL Inspection

Key Takeaways

What Scarf Buyers Need to Know About AQL Inspection

AQL (Acceptable Quality Level) is the maximum percentage of defects considered acceptable as a process average — AQL 2.5 means the process is considered acceptable if no more than 2.5% of units are defective on average; it does NOT mean 2.5% defective shipments are acceptable
The sampling plan (lot size → sample size → accept/reject numbers) is derived from ISO 2859-1 Table II-A at General Inspection Level II — this is the standard table used by virtually all third-party inspection bodies for apparel and scarf inspection
Defects are classified into three levels: Critical (safety risk), Major (functional or significant aesthetic failure), and Minor (slight imperfection) — different AQL levels typically apply to each: 0 for critical, AQL 2.5 for major, AQL 4.0 for minor
“AQL 0.0 for critical defects” means zero critical defects are accepted — if any critical defect is found in the sample, the lot is rejected regardless of the sample size or the number of critical defects; it does NOT mean zero units are inspected for critical defects
Third-party pre-shipment inspection should be commissioned on any order over 200 pieces, any new factory, and any new product — self-inspection by the factory against which the inspection is conducted is not a neutral quality verification

What AQL Means — Definitions and Statistical Basis

Acceptable Quality Level (AQL) is a statistical quality concept defined in ISO 2859-1 as “the quality level that is the worst tolerable process average when a continuing series of lots is submitted for acceptance sampling.” In practical terms, AQL 2.5 applied to an apparel programme means: if the factory produces a long-running series of shipments with a true average defect rate of 2.5%, the sampling plan will accept those shipments approximately 95% of the time (the “producer’s risk” = 5%). If the true defect rate rises above 2.5%, the sampling plan increasingly rejects shipments.

Critically, AQL is a statement about the process average — not about any individual shipment. A shipment with 4% defective units might still be accepted under an AQL 2.5 plan if the random sample drawn happens not to include the defective units in numbers that trigger rejection. Conversely, a shipment with 1% defective units might be rejected if the sample happens to contain several defects. This statistical reality is not a deficiency of the system — it is the correct understanding of how probability-based sampling works. AQL sampling is a risk management tool, not a 100% guarantee of quality.

The AQL value specified in a purchase order must be accompanied by the inspection level (typically General Inspection Level II) and the defect classification that defines what counts as a defective unit. Without all three parameters, the AQL number alone is not a usable specification.

Common AQL Levels — When to Use Which

AQL level selection depends on the consequence of a defect reaching the consumer and the commercial risk level of the programme.

0.0

For Critical defects — safety hazards. Zero tolerance: any critical defect in the sample triggers lot rejection.

1.0

For Major defects in premium programs — tighter than standard; used for luxury and high-visibility products.

2.5

For Major defects — standard for most apparel and scarf programs; most common setting in fashion buying offices globally.

4.0

For Minor defects — more permissive; minor imperfections that do not affect function or visible aesthetic are accepted at a higher rate.

AQL 2.5 Sampling Plan — ISO 2859-1, General Inspection Level II

The table below gives the sample size, acceptance number (Ac), and rejection number (Re) for common scarf order lot sizes, derived from ISO 2859-1:1999 Table II-A at AQL 2.5, General Inspection Level II. If the number of defects found equals or exceeds Re, the lot is rejected.

Lot Size (pcs)	Code Letter	Sample Size	Accept (Ac)	Reject (Re)	Example: 1200 pcs lot → inspect 125 pcs
2–8	A	2	0	1	Inspect all 2 units; 0 defects accepted; 1 defect = reject
9–15	B	3	0	1	Sample 3 of 15; 0 accepted
16–25	C	5	0	1	Sample 5 of 25; 0 accepted
26–50	D	8	0	1	Sample 8 of 50; 0 accepted
51–90	E	13	1	2	Sample 13; 1 defect accepted, 2 = reject
91–150	F	20	1	2	Sample 20 of 150; 1 defect accepted
151–280	G	32	2	3	Sample 32 of 280; 2 defects accepted, 3 = reject
281–500	H	50	3	4	Sample 50 of 500; 3 accepted, 4 = reject
501–1,200	J	80	5	6	Sample 80 of 1000; 5 defects accepted, 6 = reject
1,201–3,200	K	125	7	8	Sample 125 of 1200 lots; 7 accepted, 8 = reject
3,201–10,000	L	200	10	11	Sample 200 of 5000; 10 accepted, 11 = reject
10,001–35,000	M	315	14	15	Standard for large promotional or basic scarf runs

Important: The AQL 2.5 sampling plan applies to Major defects. For Critical defects, AQL = 0 (any critical defect found = lot rejected). For Minor defects, AQL 4.0 is typically applied — consult ISO 2859-1 Table II-A at AQL 4.0 for the corresponding Ac/Re numbers, which are more permissive than the AQL 2.5 column.

Defect Classification for Scarves

Defect classification for scarves follows the three-level system defined in ISO 2859-1 and applied in practice by international third-party inspection bodies. Classification determines which AQL level applies and what action is triggered.

Critical

Safety Hazard

AQL = 0 (zero tolerance). Any critical defect found in the sample triggers immediate lot rejection regardless of defect count. These defects pose a risk of injury or regulatory non-compliance.

Sharp metal parts (broken needle fragment in scarf body)
Strangulation risk from excessively long cords or drawstrings on children’s scarves exceeding EN 14682 limits
Flammability failure — fabric that ignites in the flammability test at a rate above the regulatory limit
Confirmed presence of restricted chemical substances (azo dyes, formaldehyde above REACH limits)
Choking hazard: small button or decorative element attached insecurely that detaches under pull-force test
Asbestos or other prohibited fibre content detected in fibre composition test

Major

Functional / Significant Aesthetic Failure

AQL = 2.5 (standard for fashion). A major defect significantly affects the use, saleable appearance, or marketability of the product. Visible at retail distance (approximately 1 metre).

Wrong colour — shade outside agreed ΔE tolerance vs approved standard
Pattern mismatch at seams exceeding 5 mm (misaligned jacquard or print repeat)
Missing fringe section — visible gap in fringe at scarf end
Dimension outside specified tolerance band — length or width outside PO spec
Sewing defect visible at 1 m: broken seam, open hem, skipped stitches in visible area
Wrong label (care symbol, fibre content, country of origin mismatch to approved)
Significant pilling visible on surface (Grade 1–2 on Martindale pilling scale)
Visible stain, oil mark, or mould on fabric surface
Print misregistration >3 mm visible in normal use position

Minor

Slight Imperfection

AQL = 4.0 (more permissive). A minor defect is a departure from specification that does not materially affect function, saleable appearance, or fitness for purpose. Not visible at 1 metre under normal retail conditions.

Slight shade variation within the agreed ΔE tolerance (e.g. ΔE 1.8 vs limit of 2.0)
Thread end not trimmed flush — protruding <5 mm in non-visible area
Minor fringe length variation within ±1 cm tolerance
Small tension irregularity in knit fabric visible only under close examination (no structural effect)
Slightly off-centre label within 3 mm of specified position
Surface fluff or fibre shed on synthetic — not indicative of fabric defect
Minor crease from packing — recovers on steaming or shaking

Scarf-Specific Defect Classification Table

Reference table mapping common scarf production defects to classification level and the applicable AQL column for inspection purposes.

Defect	Classification	AQL Applied	Pass/Fail Trigger
Broken needle fragment in fabric	Critical	AQL 0 — zero tolerance	Any 1 found = lot rejected immediately
Cord length on children’s scarf exceeds EN 14682	Critical	AQL 0	Any 1 found = lot rejected
Wrong color — outside ΔE tolerance	Major	AQL 2.5	Per ISO 2859-1 Table II-A, Ac/Re by lot size
Dimension outside PO tolerance (e.g. length <177 cm on 180±3 spec)	Major	AQL 2.5	Per ISO 2859-1 Table II-A by lot size
Pattern mismatch >5 mm at seam	Major	AQL 2.5	Per sampling table
Wrong or missing care label	Major	AQL 2.5	Per sampling table
Open seam or broken stitch visible at 1 m	Major	AQL 2.5	Per sampling table
Fringe length outside ±1 cm tolerance	Major	AQL 2.5	Per sampling table
Minor shade variation within ΔE tolerance	Minor	AQL 4.0	Per ISO 2859-1 Table II-A at AQL 4.0
Thread end <5 mm in non-visible area	Minor	AQL 4.0	Per sampling table at AQL 4.0
Packing crease — recovers on steaming	Minor	AQL 4.0	Per sampling table at AQL 4.0

Pre-Shipment Inspection Process

A properly conducted pre-shipment inspection for scarves follows a documented process that covers four elements: lot verification, random sampling, defect inspection, and dimensional and labelling checks. The inspection must be conducted by a qualified third-party inspector or buyer’s quality representative — factory self-inspection against the same shipment is not an acceptable substitute for an independent inspection because the inspector is not neutral relative to the outcome.

Lot verification: the inspector verifies that the total production quantity declared matches the actual packed quantity, and that all carton markings correspond to the purchase order. Random sampling is drawn from across the production run — not from the most easily accessible cartons — to ensure the sample represents the full lot. The sample is selected according to the ISO 2859-1 General Inspection Level II table based on the declared lot size.

Defect inspection is conducted under calibrated D65 daylight lighting (minimum 1000 lux on the inspection surface), with each piece unfolded and inspected across its full surface. Defects found are recorded by type and classification on the inspection report. Dimensional checks are conducted on a sub-sample of 10–15 pieces per the measurement protocol (flat, zero tension, post-wash if required). Labelling is checked against the approved label artwork on every inspected piece — label content, placement, stitching, and legibility.

The inspection report documents: lot size, sample size, code letter, defects found by type and classification, accept/reject result per AQL level, dimensional measurement results, and any other observations. The report is the documentary evidence for acceptance or rejection of the shipment and must be retained by the buyer.

Common Misunderstandings

Misconception 1

“AQL 2.5 means we accept up to 2.5% defective products in every shipment.”

The Technical Reality

AQL 2.5 is a process quality level, not a per-shipment defect allowance. It means the sampling plan is designed to accept lots from processes running at a 2.5% average defect rate approximately 95% of the time. An individual shipment with 2.5% defective units may be accepted or rejected depending on which specific units appear in the random sample. Conversely, a shipment with 4% defective units may be accepted in a single sampling event if the defective units happen not to fall in the sample. AQL sampling is a risk-management tool for a production process, not a pass rate for individual shipments. Buyers who communicate “2.5% defective is acceptable” to their factories are creating the wrong incentive — the goal is a process that consistently produces at or below the AQL level, resulting in shipments that are accepted consistently.

Misconception 2

“AQL 0.0 for critical defects means only zero critical defects are inspected — so it’s not a useful check.”

The Technical Reality

AQL 0 for critical defects does not mean zero units are checked. It means that the standard sample size (derived from the lot size at General Inspection Level II) is inspected for critical defects, and the acceptance number is zero — meaning that if even one critical defect is found in the sample, the entire lot is rejected. This is a much tighter standard than AQL 2.5 for the same sample size: at AQL 2.5 for a 1,200-piece lot, 7 major defects are accepted; at AQL 0 for critical defects, 0 critical defects are accepted from the same 125-piece sample. The combination of the standard sample size with Ac = 0 produces a rejection probability that rises steeply with any increase in the true defect rate — even a very small critical defect rate has a high probability of producing a rejected shipment under AQL 0.

Buyer Decision Notes — When to Commission AQL Inspection

Always Commission Inspection

Any order over 200 pieces from any source factory
First order from a new factory — no track record to rely on
First production of a new product design or construction type
Any order containing children’s scarves (critical safety risk of cord and small parts)
Reorders where prior shipments had borderline or failing inspection results
Shipments travelling by sea with >4 week transit — condition at port of loading matters
Large seasonal programs where failed lots cause missed retail windows

Lower Priority (But Recommended)

Very small orders (<100 pcs) from long-established supplier with 5+ passed inspections — risk-based decision
Repeat reorders of unchanged product from factory with consistent passed inspection history
Sample orders or development samples (inspection at PP or pre-production stage is more relevant)
Display-only or no-resale products where consumer safety and regulatory compliance are not the primary risk

Quality Risks in AQL Inspection Implementation

Undefined defect classification: The most common failure in AQL inspection programs is conducting an inspection without a pre-agreed defect classification list. An inspector who classifies a thread end as a major defect (when it should be minor) produces a higher rejection rate than the specification intends; an inspector who classifies a colour difference outside the ΔE tolerance as minor produces a lower rejection rate than appropriate. Defect definitions and classification levels must be documented in the inspection brief before inspection commences, not negotiated at the factory on the inspection day.

Sampling from accessible cartons only: Random sampling requires the sample to be drawn from across the full lot, including cartons at the back of the packing area, at the bottom of pallet stacks, and from different production days where the lot spans multiple production sessions. Sampling only from the most accessible cartons is not random and typically produces a biased (too optimistic) quality estimate. The inspection brief should specify that cartons must be made available randomly by the inspector, not selected by the factory.

Lighting conditions: Visual inspection of fabric defects — shade variation, weaving faults, surface pilling, staining — requires adequate calibrated lighting. Inspection under tungsten or mixed-source factory floor lighting will miss shade-related and surface defects that would be clearly visible under D65 daylight. Minimum 1000 lux at the inspection surface under D65-equivalent lighting is the standard requirement.

Standards & Technical References

ISO 2859-1:1999 — Sampling procedures for inspection by attributes — Part 1: Sampling schemes indexed by acceptance quality limit (AQL) for lot-by-lot inspection; the primary standard defining all sampling tables, code letters, and acceptance/rejection criteria used in this guide
ANSI/ASQ Z1.4-2003 (R2018) — Sampling Procedures and Tables for Inspection by Attributes; the US equivalent of ISO 2859-1, producing identical sampling results; referenced by many US-market buying offices
ISO 105-J01 — Textiles: Methods for colour measurement; referenced for the spectrophotometric assessment of shade defects in colour inspection
ISO 3635 — Size designation of clothes; referenced for the dimensional measurement protocol applied during AQL inspection
Third-party inspection services: SGS, Bureau Veritas, Intertek, and QIMA are among the internationally recognised inspection bodies that conduct AQL inspection per ISO 2859-1 for scarf and apparel shipments from China and other production markets

Related Technical Guides

Explore connected quality and compliance topics that integrate with AQL inspection in scarf procurement.

See this standard applied in production: WeaveEssence factory technical records and production specifications demonstrate pre-shipment inspection procedures conducted per ISO 2859-1 General Inspection Level II, with documented defect classification lists agreed prior to inspection and inspection reports archived per shipment. Buyers integrating these parameters into purchase orders typically achieve more consistent batch outcomes. ← Tech Hub Index