Workmanship criteria in fabric inspection: part 2

Key takeawaysCalibrate severity per article by judging each fault in its finished end use, since identical marks carry different consequences in shirting versus canvas. · Judge combined effects where individually minor faults interact into visible bands or patterns the customer will notice. · Choose rework for confined correctable faults and rejection for pervasive or irrecoverable ones, using distribution data as evidence. · Ground severity practice in acceptance sampling logic, where thresholds are agreed before inspection ^[1]. · Follow the acceptance-test pattern of demonstrating conformity against criteria before release ^[2].

The first instalment of this topic set out the standard workmanship criteria: which weaving, dyeing, and finishing faults inspectors record and how severity grades organise them. This second part addresses the questions buyers ask once the criteria are understood: how to calibrate severity for different articles, how inspectors judge faults that interact, when rework beats rejection, and how repeated findings convert into supplier accountability.

These are judgement questions rather than definition questions, and they resist one-size answers. A slub that is minor in furnishing canvas is major in fine shirting; a shade drift correctable by re-dyeing in one article is irrecoverable in a coated fabric. The skill lies in applying stable principles to varying articles, and in documenting each application so that verdicts stay consistent across inspectors, seasons, and mills.

What follows is that applied layer: calibration method, combined-effect reasoning, the rework decision tree, and the accountability routine that turns repeated findings into measurable mill improvement.

Calibrating severity to the article

Severity grades describe commercial consequence, and consequence lives in the finished product rather than in the roll. Calibration therefore starts from the retail shelf and works backwards: imagine the fault in a sewn shirt, a curtain panel, or an upholstery cushion under store lighting, and ask whether a customer would return it, notice it, or never see it. Holes, tears, and strong chemical odours fail in nearly every end use; prominent dye streaks fail wherever appearance sells the product; faint texture irregularities pass wherever the article's own design carries stronger visual interest. Writing these judgements into the article's defect list, with one example photograph per grade, gives every inspector the same scale.

Recalibrate whenever the article changes substantially. A mill that mastered quilting-weight canvas will not automatically master voile; the lighter construction reveals faults the heavier one absorbed, and severity grades must shift accordingly. Buyers launching new weights or constructions should review the defect list before the first bulk inspection rather than discovering mid-season that criteria inherited from a different article condemn goods customers would have accepted — or pass goods customers will return.

Combined effects and borderline judgement

Individual defect counts sometimes understate what the customer will see. Ten individually minor streaks aligned across a metre of cloth read as one visible band; scattered faint spots clustering on one garment panel read as a stained panel; slight shade drift plus slight texture variation together read as a quality step-change even where each alone sits within tolerance. Competent inspectors flag these interactions explicitly, recording the combined impression alongside the individual counts so the verdict reflects the customer's eye rather than merely the tally sheet.

A disciplined fabric inspection fabric inspection documents combined-effect reasoning in words, not just in numbers, stating which faults interacted, over what area, and why the interaction lifts the finding above its component grades. This transparency protects both sides: the buyer sees the reasoning that downgraded the goods, and the mill can address the pattern rather than contesting isolated counts. Sampling theory supports this practice structurally, since acceptance decisions ultimately rest on whether the lot meets agreed thresholds understood as commercial fitness, not as arithmetic alone ^[1]. Borderline lots judged this carefully produce fewer disputes than clear-cut lots judged opaquely.

The rework-versus-rejection decision

Use the decision logic below when the report lands. The governing principle is distribution: confined faults invite correction, pervasive faults invite rejection, and the report's spread data tells you which you face.

Accountability that improves mills

Repeated findings test the commercial relationship more than any single failure. The accountability routine that works is staged and documented: first failure triggers correction plus re-inspection at the mill's coordination cost; repetition of the same fault triggers a management review covering root cause, process change, and preventive evidence; and a pattern across articles triggers allocation review, where future volume follows demonstrated correction. Each stage is recorded with dates, so consequences visibly track performance rather than mood.

This routine mirrors acceptance-test discipline for capital goods, where demonstration of conformity precedes release and repeated demonstration failure reshapes the supply relationship ^[2]. Applied to fabric, it means mills learn that inspection findings predict order volumes, which concentrates managerial attention far more effectively than complaints. Buyers who operate the routine consistently report the same outcome: fewer repeated findings within two seasons, because mills redirect their best supervision toward the customers whose consequences are predictable.

Keeping criteria stable across inspectors and seasons

1. Freeze the defect listThe article defect list with severity grades, example photographs, and tolerance numbers, acknowledged in writing by the mill before bulk production.
2. Control the referencesSealed bulk and shade references retained under controlled storage, replaced only through a documented re-approval that records why the old reference changed.
3. Calibrate jointlyOne calibration session per season in which the buyer, the inspector, and the mill review borderline findings from the previous season and confirm grading alignment for the next.
4. Score and reviewSeason scorecards per mill tracking major rate, shade consistency, and correction closure speed, reviewed together with one improvement target set per cycle.

Frequently asked questions