Sampling Limit
Statistical probability defines a buyer risk threshold where four defective units per one hundred items passed for shipment represents the upper limit of acceptable non-conformance. This measurement, known widely as aql 4.0, dictates the frequency of rejection for a production lot based on predetermined audit intensities. Garment inspectors apply this value to visual or functional defects identified during a final random check of packaged goods.
The logic rests upon the principle that a batch contains a predictable frequency of errors while remaining commercially viable for retail distribution.
Testing Mechanism
Operators pull a fixed quantity of units from a total production run to perform this evaluation. A table dictates the exact number of pieces to inspect based on the lot size and the chosen level of inspection severity. Inspectors count every detected defect found within that extracted pool and compare the total against the rejection number defined by the statistical plan.
If the count of non-conforming items stays beneath this target, the entire batch meets the quality requirement. Rejection follows whenever the observed error rate exceeds the allowance, forcing a hundred percent screening of the remaining stock or a full return of the goods to the finishing department.
Production Verification
Textile mills and apparel factories monitor these percentages to adjust their internal quality control settings before a shipment leaves the facility. Stable processes maintain a defect rate consistently below the maximum threshold, while fluctuating results indicate a breakdown in stitch density, fabric uniformity, or dyeing precision. Auditors verify these figures at the point of packing to ensure the finished goods match the agreed technical specifications.
Reliable data at this stage prevents the shipment of faulty garments that would otherwise incur high logistical costs during reverse supply chain operations.
Threshold Consequence
Exceeding the specified allowance initiates a secondary review where the buyer decides if the identified flaws warrant a total loss of the lot or a negotiated discount. This risk management approach operates on the assumption that absolute perfection in large-scale manufacturing remains physically unattainable. Establishing a mathematical bound protects the procurement budget from overwhelming failure rates while acknowledging the reality of mass production variance.
Final inspection data provides the evidence required to hold manufacturing partners accountable for consistent output quality.