Research libraryTraceable by Design
Chapter 06 of 14 · Part II · Draft web edition
Physical stock, processing yield and eligible claims need related but separate accounts.
The cloth is woven from this chapter's 4 exhibits: each weft is one exhibit's own data, normalised; each warp group is a section. The probe reads the exhibit's own value where it crosses. Procedural visualisation, not a measurement.
Read-out
Illustrative training case, not a GFT result. Opening 100 + receipts 900 - good output 800 - recoverable transfer 100 - recorded loss 20 = modelled closing 80 kg. Returned material and dispatches are zero in this case. The recovery transfer leaves this account but remains identified elsewhere.
Unit: kg on one declared illustrative basis. Status: illustrative. Sources: S03, S06. Illustrative figures are teaching examples, not measured factory results.
Overture · 1 min
Reconciliation should stop a claim before it exceeds the material and eligibility that its records can support. Green Future Tech Limited (GFT), Hong Kong SAR, therefore treats three accounts as distinct: a physical ledger asks where material is, a transformation account explains how inputs became outputs, and a claims ledger records which attributes remain available for allocation. They must agree through explicit rules, not through a single balance that silently substitutes for all three. A warehouse can hold cotton whose origin evidence is inadmissible. A scheme account can hold eligible quantity without proving which fibres sit in a particular garment. Reconciliation makes those differences visible and gives a release reviewer a reason to stop.
This chapter specifies that control for GFT's proposed Cameroon textile work. It is a design for a later pilot, not a report of an operating factory, certified supplier, completed reconciliation or realised saving. All named identifiers, quantities, conversion factors and warning bands below are illustrative teaching inputs. They do not describe a GFT result or an industry tolerance. Scheme requirements belong to their stated editions and scopes. This chapter is not investment, legal or tax advice.
6.1
Related · chapter 018 paragraphs step it
Read-out
Illustrative seven-stage cotton-to-garment custody chain structured using S03 material tracking principles and S10 visibility event semantics. Represents a hypothetical control boundary, not an active GFT supply chain.
Unit: . Status: illustrative. Sources: S03, S10. Illustrative figures are teaching examples, not measured factory results.
Read this exhibit in its chapter ↗A physical balance is meaningful only after its account boundary is fixed. The account card should name the site, material class, custody model, stock locations, opening time, closing time and measurement basis. It should also state whether work in progress and recoverable material sit inside or outside the account. Seed cotton, lint, yarn, fabric and garments cannot share an unqualified kilogram balance. Their forms, additives and units differ. The Content Claim Standard overview expressly connects volume reconciliation with documentation for receiving, processing and transfer, while the opened Aid by Trade Foundation standard requires ongoing reconciliation by lot or batch within its own scope.[S03][S06]
For an eligible processing account, GFT would use C_model = O + R + U - D - Q - W - L. Here O is opening stock, R is admitted receipts, U is returned material accepted back into the same account, D is dispatches, Q is good output transferred out, W is recoverable material transferred to a separate recovery account, and L is recorded loss. C_model is the modelled closing quantity. The equation does not estimate the physical count. It predicts what the count should be if every flow is complete and measured on the same basis.
The sign of a flow comes from the account card, not from the document title. A supplier return is an outward movement when the original receipt remains in the account history. A customer return is an inward movement only when the material has actually re-entered the defined boundary and its identity has been accepted. A transfer to a subcontractor is not process consumption. It is a movement to another custody location, with a corresponding account and return or output linkage. These distinctions prevent a dispatch and a transformation input from removing the same quantity twice.
Cut-off discipline matters as much as the formula. A receipt entered after the close but physically present before it creates a timing exception, not free stock. A late message must retain its physical event time and later recording time. GS1's event model separates the observation of a business step from the systems that exchange it, and its vocabulary gives parties shared meanings for steps and dispositions.[S10][S11] Reconciliation should use those meanings to include each physical event once, in the correct period, without treating database arrival as the movement itself.
The illustrative balance in Exhibit 24 starts from disclosed training quantities. Recoverable material is subtracted because it leaves this account for a separately identified recovery account, not because it vanishes. A different account boundary would require a different presentation.
The waterfall accounts for the modelled flow, but it does not prove the physical count or the admissibility of any origin claim. Those are separate tests. The reviewer must still inspect the count record, scale basis, material identifiers, event references and evidence supporting the claim class. A balanced spreadsheet can be arithmetically correct and commercially unsupported.
6.3
Read-out
Illustrative transformation mass balance: 600 kg documented source A and 400 kg documented source B enter a 1,000 kg spinning laydown and allocate to 820 kg combed yarn, 130 kg recoverable comber noil and 50 kg process loss. Quantities are hypothetical teaching inputs under S10 transformation semantics and S02 claim boundaries, not empirical yield claims or GFT factory measurements.
Unit: kg (illustrative mass). Status: illustrative. Sources: S02, S10. Illustrative figures are teaching examples, not measured factory results.
Read this exhibit in its chapter ↗6.3 · Introduction
A conversion factor is an output of measured production, not a licence to create eligible output.
Exhibit 20, chapter 05: Documented source A (600 kg) to Spinning laydown (1,000 kg)
6.4 · Work in progress belongs to the period account
A period close rarely coincides with empty machines.
Exhibit 20, chapter 05: Documented source A (600 kg) to Spinning laydown (1,000 kg)
6.5 · Moisture and finishing can imitate excess production
Mass changes that have an explained physical basis must not be read as new cotton.
Exhibit 20, chapter 05: Documented source B (400 kg) to Spinning laydown (1,000 kg)
6.3 · the bodyThe film above follows these sub-steps.
A conversion factor is an output of measured production, not a licence to create eligible output. For a completed process and a declared material basis, GFT would calculate good-output yield as Y_good = M_good / M_in. Separate ratios should describe recoverable outputs and recorded losses. Their purpose is to explain the observed process, including material still in production, rather than to force every batch towards a benchmark. The opened Aid by Trade Foundation standard calls for actual input and output records, physical inventory and comparison of actual with expected waste factors in its volume-reconciliation procedure.[S06]
The denominator must match the numerator. A lint-to-yield ratio cannot use seed-cotton receipts without a measured ginning link. A fabric-to-garment calculation cannot switch silently between kilograms, metres and garment counts. Where several input lots feed one campaign, the transformation record must preserve the eligible input set and its quantities. GS1's TransformationEvent explicitly records the relationship between inputs and outputs, including the possibility that any listed input contributed to each output.[S10] That relationship supports calculation, but it does not justify a more precise physical origin than the process can establish.
A period close rarely coincides with empty machines. Material may be opened, partly processed, awaiting inspection or held for rework. The account therefore needs identified work-in-progress quantities and states. Completed-output yield should use completed output and the inputs consumed to produce it. A separate movement balance should explain changes in work in progress. Combining the two without adjustment can make a process appear to gain material merely because opening work was completed, or lose material because closing work has not yet emerged as output.
The practical record is a campaign bridge. It lists opening work, new inputs, completed good output, recoverable outputs, recorded losses and closing work. Each quantity carries the same material and moisture basis, or an explicit measured conversion. The reviewer checks that opening work equals the prior period's accepted closing work and that a batch cannot be counted both as closing work and as completed output. Any estimated work-in-progress quantity remains labelled as an estimate, with its method and approver, rather than becoming a precise observed mass.
Mass changes that have an explained physical basis must not be read as new cotton. Consider an illustrative constant dry mass of 900 kg. At an illustrative moisture share of 10 percent, its wet mass is 900 / 0.90 = 1,000 kg. At an illustrative moisture share of 20 percent, its wet mass is 900 / 0.80 = 1,125 kg. The 125 kg difference is water in this constructed case, not additional fibre. A wet-mass comparison without the moisture basis could falsely present that difference as excess output.
Consider an illustrative constant dry mass of 900 kg
illustrative
Finishing creates a similar risk. If an illustrative 800 kg fabric receives 40 kg of documented non-cotton finishing additions, the resulting 840 kg as-is fabric mass does not contain 840 kg of cotton. The cotton account, additive account and total product-mass account must retain their separate quantities. The correct treatment depends on the process and claim rule, but the arithmetic boundary is constant: an additive cannot inherit cotton eligibility merely because it changes the weighed total.
If an illustrative 800 kg fabric receives 40 kg of documented non-cotton finishing additions
illustrative
Expected yields remain useful as questions. A shift away from an established range can prompt a check of scale condition, input quality, machine settings, waste handling or record completeness. It cannot prove substitution or error by itself. Nor should a favourable yield automatically increase the claims ledger. Eligibility follows the chosen scheme's rules and the evidence attached to the actual input and output, not the mere existence of a calculated gain.[S02][S03]
6.6
The claims ledger records a right to allocate a defined attribute under specified rules. It is not a second name for physical inventory. ISEAL describes custody models as transferring materials and associated attributes, while Cotton made in Africa's current overview describes Mass Balance as quantity-based tracking with equivalent verified volumes accounted for.[S02][S07] Neither proposition says that a positive account balance identifies the physical fibres in one garment. The claim statement must retain that limit.
GFT would write the account as E_close = E_open + E_admit + E_reverse - E_alloc - E_expire - E_downgrade. E_open is the accepted opening eligible balance, E_admit is quantity admitted after purchase and evidence review, E_reverse is a permitted cancellation reversal, E_alloc is quantity allocated to released claims, E_expire is quantity no longer eligible, and E_downgrade is quantity moved to a narrower claim account. The unit may be a scheme-defined equivalent quantity rather than observed mass. Each term needs a rule reference and transaction evidence.
Physical receipt does not automatically create eligibility. The admission review should test supplier and site scope, material class, custody model, transaction evidence, quantity basis and any required confirmation. The ISO public abstract warns that a chain-of-custody framework alone is not intended to make or verify a product claim.[S01] The same limitation applies here: an account structure cannot decide admissibility without the evidence and rules that give an entry meaning.
A rejected or pending quantity remains visible. It should appear as PENDING, INELIGIBLE or UNKNOWN, as appropriate, rather than being entered as zero eligible stock and forgotten. Zero can be a calculated balance. Unknown means that the required fact has not been established. This distinction stops an evidence gap from disappearing inside an apparently clean close.
The illustrative registers in Exhibit 25 deliberately produce different closing balances. They show why a reviewer must reconcile the relationship between accounts rather than expect numerical identity. The figures do not represent any scheme's conversion rule.
Read-out
| Purpose | Where is the material? | Which attribute can still be allocated? |
|---|---|---|
| Opening | 100 kg | 40 kg equivalent |
| Admitted increase | 900 kg received | 850 kg equivalent admitted |
| Released or allocated | 800 kg good output | 700 kg equivalent allocated |
| Other decreases | 100 kg recovery transfer; 20 kg loss | 50 kg equivalent expired or downgraded |
| Permitted reversal | No return in this case | 30 kg equivalent cancellation reversed |
| Closing | 80 kg | 170 kg equivalent |
| Release evidence | Count, basis, genealogy and custody | Eligibility, rule version and unique allocation |
Illustrative training registers, not an operating account or a scheme conversion example. Physical closing: 100 + 900 - 800 - 100 - 20 = 80 kg. Claim closing: 40 + 850 + 30 - 700 - 50 = 170 kg equivalent. The difference is not itself a nonconformity; the account boundaries, rule versions and transaction links must explain it.
Unit: illustrative kg and scheme-defined kg equivalent. Status: illustrative. Sources: S02, S03, S06. Illustrative figures are teaching examples, not measured factory results.
A difference between the two closing quantities is not automatically an error. It may follow from their different definitions, periods or scheme rules. It becomes an exception when the declared relationship between the accounts cannot explain it. For example, an allocation that cites no admitted input, uses an obsolete conversion rule or exceeds the available balance is unsupported even if the physical warehouse is full. Conversely, destroying physical material does not necessarily tell the reviewer how every scheme requires the corresponding attribute account to be adjusted. That adjustment needs its own rule and approval.
The minimum allocation record should identify the claim account, scheme and edition, custody model, site, period, input transactions, transformation or conversion rule, output lot, quantity, unit, allocation time and release reference. It should also identify the evidence approver and any cancellation that later affects the allocation. These fields let another person reconstruct why the quantity was available and where it went.
Interlude · the cloth half-woven
Woven: sections 6.0 to 6.6Still to weave: 6.9 to 6.13
6.9
Related · chapter 147 paragraphs step it
Read-out
Illustrative GFT decision rule. A No at any gate holds the affected claim and creates a remediation record. Independent acceptance covers only the reviewed scope and evidence cut-off. This is not a completed implementation or certification decision.
Unit: illustrative decision sequence. Status: illustrative. Sources: S03, S06, S10, S14, S27. Illustrative figures are teaching examples, not measured factory results.
Read this exhibit in its chapter ↗Duplicate claiming is a release failure, not merely a reporting defect. A nightly reconciliation can discover that two sales consumed the same eligible quantity, but by then both buyers may hold an unsupported statement. The control must therefore execute when an allocation is proposed. The opened Aid by Trade Foundation standard prohibits the same CmiA or RCS volumes being claimed under both that programme and another certification or sustainability programme, within its stated scope.[S06] It is a scheme-specific rule, not a complete definition of every conflict that a factory should prevent.
GFT would give each allocation a unique identifier and bind it to one claim account, one output set and one released claim statement. Before acceptance, the release transaction should confirm that the input allocation is active, the output genealogy is admissible, the conversion rule is current for the intended claim, the available balance is sufficient and no incompatible allocation already exists. All checks and the debit should succeed or fail together. A separately generated label or passport entry must not become visible before that transaction commits.
A retry after a network failure should return the original allocation result, not create a second debit. This requires an idempotency key supplied by the requesting process and a stored response tied to that key. A cancellation should append a linked reversal that identifies the original allocation, the quantity reversed, the reason and the authority. It should not delete the original record. Shared event vocabulary helps parties distinguish a physical movement from its later business state, but the uniqueness rule remains an application control.[S10][S11]
Returns need the same discipline. Physical return of a garment, cancellation of an invoice and restoration of an eligible balance are different events. The last should occur only if the governing rule permits it and the complete original allocation can be identified. A downgrade also needs an explicit destination account. A later purchase must not be used to repair an earlier unsupported physical claim or to erase a sale recorded under a narrower claim.
The failed request is useful evidence. It should record the requested quantity, available balance, conflicting allocation where relevant, rule version and reviewer decision. Operations can then investigate whether the cause is a genuine shortage, a duplicated message, an incorrect conversion, an expired input or a commercial attempt to make incompatible claims. The system should not infer intent from the arithmetic.
Where staffing permits, the person who enters an adjustment should not be the only person who approves its release. Independent review is particularly important for manual balance corrections, cancelled sales and changes to conversion rules. The Content Claim Standard overview links credible content claims with verified tracking and documentation, which supports a release decision grounded in inspectable records rather than dashboard availability alone.[S03]
6.11
The residual is a diagnostic quantity, not a verdict. GFT defines it as r = C_obs - C_model, where C_obs is the observed closing quantity and C_model is the modelled closing quantity. A positive residual means that observed stock exceeds the model. Possible causes include an omitted receipt, an overstated outbound flow, a duplicated outbound record or an incorrect count. A negative residual means that observed stock is below the model. Possible causes include an omitted dispatch, an unrecorded loss, a duplicated receipt or an incorrect count. The sign convention must travel with every report.
A zero residual is narrower than it appears. It says that the observed and modelled closing quantities agree under the stated basis. It does not prove that each input was genuine, every output carried a valid claim or two offsetting errors were absent. Nor does a non-zero residual establish theft, substitution or false origin. It locates a question for investigation.
Every observed quantity has a measurement context. Scale resolution, container tare, moisture assessment, roll-length conversion, work-in-progress estimation, timing and rounding can all affect the close. The review should record which components are measured, which are estimated and which are unknown. It should not attach an invented confidence interval to the result. Where an uncertainty allowance is used, its derivation, coverage meaning and source observations must be stated.
GFT would report the residual together with an explicitly derived review allowance U, rather than present r as an exact physical truth. The comparison r against U can help prioritise investigation, but it cannot make an unsupported claim eligible. If the inputs needed to derive U are missing, the allowance is UNKNOWN, not zero. A pilot should gather repeated counts and process variation before proposing an operational warning band.
The training series in Exhibit 26 uses a fixed teaching band solely to demonstrate why the band's basis matters. It is not measured process variation and not an accepted tolerance.
Read-out
Illustrative teaching series. Residuals are 0, 12, -8, 18, -15 and 5 kg for six synthetic closes. The selected review band is plus or minus 20 kg at every close, chosen only for display. It is not a statistical interval, measured process limit, GFT tolerance or scheme requirement.
Unit: kg, illustrative observed-minus-model residual. Status: illustrative. Illustrative figures are teaching examples, not measured factory results.
All six illustrative residuals sit inside the selected band, yet that fact alone would not justify automatic release. A real review needs the band's empirical basis and must still examine systematic errors that a wide band could hide. Equally, a value outside a band is an investigation trigger, not proof of misconduct.
Scheme rules must remain separate from internal review allowances. The opened June 2025 Aid by Trade Foundation standard requires deviations of more than five percent from expected volumes to be investigated and communicated for volume correction within its scope.[S06] That obligation must not be universalised to another scheme, and it does not mean that a smaller deviation is necessarily immaterial. The illustrative 20 kg band above has no relationship to that requirement.
The illustrative 20 kg band above has no relationship to that requirement
illustrative
6.13
Missing receipt
Missing receipt: Physical residual, observed minus modelled (kg) 200, Unsupported claim quantity (kg equivalent) 0 · illustrative
A close is complete only when another competent person can reconstruct it without asking the preparer which hidden adjustment made it balance.
Duplicate receipt
Duplicate receipt: Physical residual, observed minus modelled (kg) -200, Unsupported claim quantity (kg equivalent) 200 · illustrative
Duplicate dispatch
Duplicate dispatch: Physical residual, observed minus modelled (kg) 150, Unsupported claim quantity (kg equivalent) 150 · illustrative
Late return
Late return: Physical residual, observed minus modelled (kg) 60, Unsupported claim quantity (kg equivalent) 0 · illustrative
Wrong conversion
Wrong conversion: Physical residual, observed minus modelled (kg) 0, Unsupported claim quantity (kg equivalent) 90 · illustrative
Recovery misclassified
Recovery misclassified: Physical residual, observed minus modelled (kg) 0, Unsupported claim quantity (kg equivalent) 100 · illustrative
The method behind it: GFT note.
Illustrative arithmetic from the base closing balance of 80 kg. Missing receipt 200: model -120, residual +200, no claim allocation. Duplicate receipt 200: model 280, residual -200, duplicate eligibility 200. Duplicate dispatch 150: model -70, residual +150, duplicate claim 150. Late return 60: model 20, residual +60, no claim allocation. Wrong conversion on 900 eligible kg: 900 x (0.80 - 0.70) = 90 kg unsupported, physical residual 0; both factors are hypothetical. Recovery misclassification: 100 kg incorrectly admitted, physical residual 0. No factor or effect is a scheme benchmark.
Unit: kg, illustrative residual and unsupported claim quantity. Status: illustrative. Sources: S03, S06, S10. Illustrative figures are teaching examples, not measured factory results.
6.13 · the bodyExhibit 27 above; its figures are illustrative where marked.
A close is complete only when another competent person can reconstruct it without asking the preparer which hidden adjustment made it balance. The reproducibility pack should contain the signed account card, opening balance reference, source-event extract, physical count records, measurement bases, claim-account entries, conversion-rule versions, rounding rules, calculations, exceptions and approvals. It should identify excluded records and explain each exclusion. A reviewer begins with those source records, calculates an independent expected result and compares that result with the preparer's work.
This is more than retaining a spreadsheet. The pack must preserve the relationship between a physical movement, its transformation genealogy and its claim allocation. EPCIS supplies a pinned way to describe transformation inputs and outputs, while the Core Business Vocabulary supplies agreed business-step and disposition meanings for exchanging events.[S10][S11] Those standards do not calculate the factory's balance. They reduce ambiguity in the records from which the balance is rebuilt.
The illustrative cases in Exhibit 27 start from the earlier balanced account, with observed and modelled closing stock both equal to 80 kg. Each case changes one stated condition. The horizontal position is the resulting physical residual, observed minus modelled quantity. The vertical position is the unsupported claim quantity created by the accompanying claim error. A zero on either axis means no effect under that particular constructed case, not safety in every operating condition.
The matrix separates defects that a stock count can reveal from defects that can survive a perfect stock count. A missing receipt changes the physical residual in this case because the actual material is counted while its inflow is absent from the model. A duplicated receipt has the opposite sign because the model expects material that does not exist. A duplicated dispatch creates a positive residual because the model removes too much stock. These are conditional signatures, not automatic diagnoses. Other combinations of errors can produce the same result.
The wrong-conversion and recovery-misclassification cases are more dangerous for claim release because they leave the illustrative physical residual at zero. Only the claims ledger and its rule lineage expose the overstatement. That is why a clean physical close cannot serve as the sole release gate. The reviewer must test eligibility, conversion and duplicate allocation even when r = 0.
GFT would apply the same investigation order to every exception. First, preserve the original close and block the affected claim allocations. Second, confirm the account boundary, cut-off, material identity and measurement basis. Third, recount or inspect the relevant physical material where practicable. Fourth, replay receipts, dispatches, returns, transformations and claim allocations from their identifiers. Fifth, test conversion versions and duplicate keys. Finally, append the evidenced correction and obtain release approval. The sequence avoids choosing whichever explanation most quickly produces a zero residual.
Corrections must identify their cause at the level actually known. MISSING_RECEIPT, DUPLICATE_EVENT, CUT_OFF_REVIEW, BASIS_REVIEW and RULE_VERSION_REVIEW are useful working states while evidence is gathered. They should not be replaced with a definitive fraud or loss category without supporting facts. If the cause remains unresolved, the correction can make the arithmetic transparent but it cannot make the original uncertainty disappear.
A repeatable specification for the educational trace-a-garment calculation should therefore accept an account card and an ordered event set, then produce separate physical, transformation and claim balances. It should expose every input, formula, rule version, exclusion, residual and warning. It should reject incompatible units and unresolved references rather than coerce them. Replaying the same event set should produce the same result, and repeating one event identifier should not create another business movement or allocation. The output is an educational calculation until real inventory and evidence are independently checked.
The release decision should end with three explicit outcomes. SUPPORTED means that the physical, genealogy and eligibility checks required for the stated claim have passed. RESTRICTED means that only a narrower statement is supported. HOLD means that the available records cannot support release. Each outcome needs a scope, reviewer, time and link to the reconciliation pack. The commercial team then receives a claim it can explain, while operations retains a precise list of what still needs correction. This turns reconciliation into an operating control: it does not merely explain the past, it determines what may be said next.
Check the record
Every knot is a source this chapter cites; select one to read its record, or filter the register by grade. Publication dates and inherited-source limits stay explicit. Read the paper's full evidence summary.
7 sources · 25 citation marks on this pageMost cited: S06, 5 times
Grade AGrade BGrade CKnot size: citations on this page
All 7 records shown
c06-S01 · Grade A
ISO 22095:2020: Chain of custody, General terminology and models
It is not intended to be used on its own to make or verify such claims.
2020-10. Register access date: 2026-10-30. Source provenance is supplied by the paper evidence register.
Open the recorded source ↗c06-S02 · Grade B
Chain of Custody Models and Definitions Guidance
Chain of custody models enable the transfer of materials and associated attributes through certified supply chains.
2025. Register access date: 2026-10-30. Source provenance is supplied by the paper evidence register.
Open the recorded source ↗c06-S03 · Grade B
The Content Claim Standard
The CCS requires clear documentation for receiving, processing, and transferring certified materials, volume reconciliation, and appropriate controls for material segregation, helping to ensure transparency.
2026-10-30. Register access date: 2026-10-30. Source provenance is supplied by the paper evidence register.
Open the recorded source ↗c06-S06 · Grade B
Transparency Standard, Version 1.0
The organisation shall ensure that CmiA/RCS volumes are not claimed under both CmiA/RCS and other certification or sustainability programme.
2025-06. Register access date: 2026-10-30. Source provenance is supplied by the paper evidence register.
Open the recorded source ↗c06-S07 · Grade B
How Do We Track Every Fibre?
Mass Balance uses quantity-based tracking instead, allowing flexibility in complex supply chains while ensuring that equivalent volumes of verified cotton are accounted for.
2026-10-30. Register access date: 2026-10-30. Source provenance is supplied by the paper evidence register.
Open the recorded source ↗c06-S10 · Grade A
EPCIS Standard, Release 2.0
The TransformationEvent captures the relationship between the inputs and the outputs, such that any of the inputs may have contributed in some way to each of the outputs.
2022-06. Register access date: 2026-10-30. Source provenance is supplied by the paper evidence register.
Open the recorded source ↗c06-S11 · Grade A
Core Business Vocabulary Standard, Release 2.0
Vocabularies are used extensively within EPCIS to model conceptual, physical, and digital entities that exist in the real world.
2022-06. Register access date: 2026-10-30. Source provenance is supplied by the paper evidence register.
Open the recorded source ↗Keep the thread · chapter 06 of 14
Move the shuttle along the rack to preview any chapter: its sections, exhibits and reading time. Draft web edition. This research is not investment, legal or tax advice; illustrative models are not measured factory results.
The paper's rack · 14 boltsBolt length: reading time
01Part I: Claims must match the chain
GFT argues that provenance becomes useful when custody controls preserve the evidence behind a saleable claim.
02Part I: Claims must match the chain
Select the custody model by the attribute that must survive, then design operations to maintain that attribute.
03Part I: Claims must match the chain
An origin claim needs a defined producer boundary and a documented link from receipts to ginned fibre.
04Part II: Every transformation needs a record
A shipment is traceable only when dispatch and receipt refer to the same material and discrepancies remain visible.
05Part II: Every transformation needs a record
Each production step must preserve enough input-output linkage to support the claim carried by its outputs.
06Part II: Every transformation needs a record
Physical stock, processing yield and eligible claims need related but separate accounts.
07Part II: Every transformation needs a record
Assurance is credible when its issuer, scope, validity and sampling limitations remain attached to the claim.
08Part III: Passports must preserve evidence
Interoperability requires common semantics, stable identifiers and a tested way to exchange corrections.
09Part III: Passports must preserve evidence
A useful passport identifies its product, states who makes each claim and reveals the limits of the supporting evidence. Proposed fields must remain distinct from future legal requirements.
10Part III: Passports must preserve evidence
Cryptography can preserve an assertion and identify its signing authority, but accepting a textile claim still requires physical evidence and an explicit trust policy.
11Part III: Passports must preserve evidence
A system survives only if responsibilities, exceptions, privacy and constrained connectivity are part of ordinary work.
12Part IV: Scale follows demonstrated control
Compare the cost of producing admissible evidence with bounded benefits rather than an assumed premium.
13Part IV: Scale follows demonstrated control
Pilot success requires adverse cases, reproducible evidence and independently reviewed release decisions.
14Part IV: Scale follows demonstrated control
GFT recommends a staged programme in which every expansion follows demonstrated control of evidence, quantities and disclosure.