Shade Bands: One Discipline from Vessel to Lay Table
The most expensive shade defect in a garment factory is invisible until assembly: a body panel from one dye batch sewn to sleeves from another, each acceptable alone, visibly different together. No inspection station catches it in time, because each piece is fine; only the combination is wrong. The defense is a discipline that runs unbroken from the dye vessel to the lay table.
Band at the vessel
It starts where variation is born. Every dye batch is assessed against the standard and assigned a shade band, A, B, C, or the mill's own scheme, and the band travels with the batch's lots from that moment: on the fabric, in the system, through every subsequent movement. A batch without a band is a decision postponed to the worst possible place, the cutting floor at lay time.
The rule at the lay table
The cutting rule is absolute and simple: one lay, one shade band. Panels cut together get sewn together, so the lay is the last moment the combination can be controlled. Rolls staged for a lay are checked by band before spreading; a roll from band B does not join an A lay to finish a marker, however convenient. This is the ply and shade discipline enforced where it is cheap, in the staging area, instead of discovered where it is ruinous, in finished garments.
The system's part
Discipline held in heads fails on night shifts. The workable version has the band as data on the lot: the dyehouse records it once, the fabric store sees it on every roll, and the cut plan groups lays by band automatically, so the default path is the safe path and mixing requires a deliberate, recorded exception. That is the pattern worth generalizing: constraints that travel with material, the same mechanism that carries certificates and compositions in the lot handbook, enforce themselves; constraints that live in instructions get tired.
Fewer bands to manage
Upstream, the dyehouse shrinks the problem before cutting inherits it: batch sizing inside machine windows and honest greige lot selection reduce batch-to-batch spread, and right-first-time work reduces the corrective batches that widen it. The chain view is the point. Shade is not a dyehouse metric or a cutting rule; it is one discipline with three floors, and it holds exactly as well as its least documented handover.