Author
Intelactory Team
Articles by Intelactory Team
From GRN to GL: Why Your Two Books Disagree
The store's ledger and the finance ledger describe the same events in different units. Wire the posting chain and month-end agreement is automatic.
Seeing the Capacity Crunch Six Weeks Out
Overload discovered on the week costs overtime and airfreight. The same overload seen six weeks early is a conversation. Booking discipline decides which one you get.
AQL, Explained Without the Mystery
What AQL 2.5 actually promises, what a sampling plan can and cannot catch, and why passing inspections is not the same as making quality.
Inline vs Endline QC: Where to Spend the Checkers
A defect at the station costs a re-stitch; at the end of line it costs unpicking; at the buyer's DC it costs the relationship. Position decides price.
A Defect By Any Other Name Ruins Your Data
Three checkers, three names, one defect. Without a shared taxonomy, quality data cannot be aggregated, compared or acted on.
Quality Holds That Actually Hold
A red tag is an opinion. A hold that blocks issue, packing and shipment in the system is a fact. What structural quality holds look like.
Liquor Ratio, OWF, and What a Batch Really Costs
Bath chemistry scales two ways: on the water and on the fabric. Keeping the two straight is the difference between costed batches and guessed ones.
Dye Lot Sizing: The Machine Sets the Window
Batches outside the vessel's working band buy one saved batch and risk several re-dyes. How to size dye lots the machine and the shade both accept.
Shade Bands: One Discipline from Vessel to Lay Table
The costliest shade defect appears at assembly, where no inspector can catch it. Shade banding is one unbroken chain across three floors.
Fiber to Carton: What Full Lot Lineage Takes
Eight links connect a carton to its fibers. Where lineage chains actually break, and the three record-keeping rules that keep them whole.
ESPR and the DPP Timeline, Read as a Factory
The EU passport rules arrive by delegated act around 2027. The factory work takes seasons and cannot be back-filled. That asymmetry is the whole argument.
Water, Energy, Carbon: Numbers Per Stage or Not at All
Factory-level averages satisfy forms and inform nobody. Per-stage resource figures come from records you already keep, and they change what you fix first.
Certificates Should Expire Loudly
A lapsed certificate is discovered by the buyer, and by then it is shipments, not paperwork. Expiry should escalate to a named owner three times.
The Audit-Ready Factory Doesn't Prepare for Audits
Audit week is wasted effort manufacturing evidence after the fact. Factories that operate on records generate their audit trail as exhaust.
Yarn Store Discipline: Counts, Cones, and Lots
Yarn hides behind three unit systems and a culture of 'it is all cotton anyway'. Three disciplines stop the leaks: explicit counts, lot identity, reconciliation.
Knit Fabric Consumption: The Math Behind the Buy
Fabric is two thirds of material cost and bought off an estimate. The GSM chain, the width dependency, and the loop that tightens the standard every style.
Carton Math, Container Math, and the Cost of Air
Slightly wrong carton math means freight bought for air or cartons bursting on the stack. All of it is knowable the day the order is confirmed.
CM Costing from First Principles
Cost of making rebuilt from its two real inputs: measured minutes and the factory's cost per minute. No copied numbers, no invisible padding.
The FOB Price Buildup, Line by Line
From CM to the offer sheet: materials, trims, washing, testing, commercial costs and margin, stacked explicitly so nothing rides for free.
Delay Letters and the Price of Surprise
Late bad news costs more than bad news. How factories that report early and specifically keep buyers that slip-prone factories lose.
The PP Sample Is a Gate, Not a Formality
What a pre-production sample must actually prove before bulk, and why sealing one late is more expensive than sealing one honestly.
Line Balancing and the Operation Bulletin
The operation bulletin is the sewing line's blueprint. How balancing distributes work, where balancing loss hides, and what a published bulletin owes the floor.
Operator Skill Matrices: Who Can Sew What
Line balance fails on paper skills. A living operator-by-operation grade matrix is the difference between a plan and a wish.
Learning Curves: The First Three Days of a Style
Every style change has a valley. Factories that plan the learning curve recover in days; factories that deny it blame the line for physics.
SMED on the Sewing Floor
Most changeover work can happen while the old style still runs. Separating external from internal setup is the cheapest capacity a factory can buy.
Line Efficiency: Five Ways Factories Get It Wrong
Earned minutes over attended minutes is a short formula with five popular corruptions. Each one makes the number lie in a different direction.
Backward Scheduling from the Ship Date
Start from the vessel cutoff and walk upstream through every stage's real duration. Where the walk lands tells you today whether the date is safe.
WIP Gaps: Where Quantities Go to Die
Cutting says 4,800, sewing says 4,650. Between departments, undocumented handovers rot counts. Documents at every gap keep the factory honest.
Absenteeism Is an Hourly Number
A monthly absenteeism percentage hides the hole in line 4 this morning. What absence really costs a balanced line, hour by hour.
The Two Prices of a Machine Breakdown
The repair bill is the small price. Interrupted production, blown balance and delivery risk are the big one, and only one of them gets budgeted.
Why factory software fails at adoption, and what the floor needs first
The graveyard is full of good-enough products the floor quietly refused. Adoption is decided by who gets value at the moment of entry, by seconds per capture, and by sequencing rollouts through trust.
One platform or best-of-breed: the integration tax nobody budgets
Each piece wins its category comparison; the system they form loses. Interfaces you own without wanting to, timing windows, identity drift, and split accountability - priced for five years.
The order you should not have taken: reading capacity before you commit
The most expensive order of the year looked wonderful in the meeting. The information that would have said no existed at the handshake - specific capacity, honest efficiency, and a backward calendar.
ERP, MES, or operating system: what a composite mill actually needs
Strip away the acronyms and there are three shapes of factory software. For a composite mill the process is a chain, so the software must be one - and the dye-batch test tells you which shape you are looking at.
The real cost of running a factory on spreadsheets
The licenses are free; the costs arrive as a reconciliation payroll, a version tax, discovery lag priced in air freight, and a knowledge mortgage held by whoever built the workbook.
How to evaluate factory software: twelve questions that expose a demo
Demos are rehearsed on data chosen to behave. Twelve specific questions - drill this number, enter our real style, time a floor entry, export everything now - whose answers are hard to fake.
Capacity planning that respects reality: three honesty rules
Capacity is finite so the plan must choose; efficiency is never silently 100 percent; a start date needs its materials. Break any of the three and the plan becomes a wish with dates on it.
The 07:30 production meeting: running a factory on numbers that drill
A production meeting works when every number on the table can be drilled to its documents on the spot. An agenda for the meeting, and the capture discipline it quietly demands.
The lot discipline handbook: six rules that make traceability real
Everything that moves is a lot; nothing moves without a document; lineage is written at the moment or never; WIP is not inventory. The unglamorous foundation under every audit and passport.
Measured SMV or typed SMV: the difference that decides your costing
Two factories quote the same polo from two SMVs: one measured last month with provenance, one typed three seasons ago. Only one of them is a number, and everything downstream inherits the guess.
Right first time in dyeing: the most expensive percentage in the mill
A dyehouse at 85 percent RFT reprocesses one batch in seven, at the bottleneck, with doubled effluent and a slipped schedule. Priced honestly, few numbers in a mill repay attention faster.
Fabric is half your cost: marker efficiency and the cutting room
Labor gets the meetings; fabric gets a typed wastage percentage. Each point of marker efficiency on a 10,000-meter order is a hundred meters that either ships as garments or leaves as waste.
What an intelligent factory operating system actually is
Most factory software records the factory. An operating system runs alongside it: one connected model where a buyer date change reaches planning, stores, and the floor in the same hour.
The loss tree: naming where a sewing line's efficiency goes
A line at 47 percent starts arguments and settles nothing. Decomposed into balancing, changeover, downtime, absenteeism, and rework, the same number becomes a queue of improvement work with owners.
Digital Product Passport readiness: a practical checklist
Seven checks that separate factories where the passport is a byproduct of normal operation from factories where every shipment becomes a data scramble.
Why time and action plans fail, and what a TNA engine changes
Most TNAs freeze at order confirmation and keep reporting dates that stopped being true. An engine keeps the plan alive: milestones carry evidence, dependencies recompute, the critical path stays visible.