AM.
Back to Overview

Designing a Material Balance for a Sorting Facility

Mandate & Focus

Design study · Fictional mechanical-sorting facility — independent design

By Design study

Fictional facility and invented quantities. Software tests check reconciliation rules only. No field data, recovery improvement, downstream recycling, AI evaluation or policy experiment is claimed.

Design feature 01

Four synthetic accounts

Design feature 02

Explicit inventory boundary

Design feature 03

No recovery yield claimed

Executive Summary

An inspectable material account separating period flows, opening and closing stock, residue and unexplained differences, with a synthetic offline reference.

Architectural Deep Dive

A fictional facility and a deliberately narrow implementation

This Design study proposes a material-account review for a fictional mechanical-sorting facility. It would connect a reporting period's opening stock, deliveries, separated material output, residue and closing stock. No real facility, client, municipality, shipment or recovery result is represented.

The implemented component is an offline reconciliation function with synthetic software tests. Receiving integrations, inventory measurements, downstream verification, operator screens and operating policies remain proposed. The component checks arithmetic and input rules; it cannot establish that a recorded load exists or that separated material was recycled.

Define the accounting boundary

Use one facility and one explicitly documented reporting window. A production record would identify the facility, start and end instants, instruments, measurement basis, record versions and accountable reviewer. The public fixture is only an invented aggregate account, with no actual site identifier or instrument record.

The proposed boundary counts external deliveries once. Movements between internal bays do not become new incoming material. Opening and closing inventories must describe the same physical scope and measurement basis. If stock categories change, the reviewer needs a documented mapping rather than an unexplained change in totals.

For this simple mechanical-sorting example, other inflows and outflows are explicitly zero. A real process might require additional streams, such as added water or measured moisture changes. Those quantities need independent evidence; they must not become convenient plugs for an unexplained discrepancy.

Reconcile a fixed synthetic account

The fixture opens with 120 kilograms of stock and receives 1,000 kilograms during the period. It records 620 kilograms of separated material leaving the boundary, 230 kilograms of residue leaving, and 270 kilograms still on site at closing. The reference stores these quantities as integer grams to avoid binary floating-point rounding in the demonstration.

Opening stock plus incoming mass equals 1,120 kilograms. Outputs plus closing stock also equal 1,120. The arithmetic therefore balances for this invented account. This is not a measured facility result, and no measurement uncertainty has been estimated.

Opening stock also creates an attribution problem: the outgoing 620 kilograms cannot automatically be assigned to the new 1,000 kilograms of deliveries. The proposed review would require cohort tracking or a separately justified attribution rule before describing a yield for those deliveries. The released function intentionally produces no recovery percentage.

Preserve the stage that the evidence supports

EPA's recycling-system overview distinguishes collection, processing and remanufacturing into new products. This study uses that stage distinction to limit its language. It does not apply US legal requirements to the fictional facility or verify any downstream manufacturer. [2]

The proposed record would distinguish a dispatch from receipt, acceptance, further processing and later rejection. Those records may arrive from different parties. The reviewer should preserve their provenance and unresolved status instead of treating an invoice or truck departure as proof of completed recycling.

SDG 12.5.1 metadata describes a national indicator with its own reporting scope and measurement considerations. This local account implements none of that national aggregation. It is preparation for an inspectable facility balance, not an official recycling indicator or a compliance calculation. [1]

Specify incomplete and discrepant cases

The function requires the declared unit to be grams and all named fields to exist. A quantity may be an explicit unknown, represented by null. Unknown values lead to an incomplete result with no numeric discrepancy, even if the other known terms happen to match.

Known quantities must be nonnegative integers. Booleans, decimal numbers, strings, negative values and unexpected fields are rejected under this narrow contract. A production adapter would need a separate reviewed procedure for instrument precision, rounding and unit conversion; none is silently inferred here.

For a complete account, the signed discrepancy is available mass minus accounted mass. Zero is balanced. A nonzero difference inside the declared demonstration tolerance is labelled within tolerance while retaining its exact value. A larger difference remains unreconciled. The sign supplies a starting point for investigation, never a diagnosis.

Propose accountable review

A future review queue would display the physical boundary, all terms, missing quantities and discrepancy together. The operator could inspect source records, request a stock check or identify a duplicated external transfer. Any correction would preserve the previous account and its reason for revision.

The proposed workflow should prevent an AI summary from renaming an unexplained gap as recovery or loss. Testing that behavior requires an implemented application and attributable evaluations. Neither a model nor such an application was run for this study.

What readers can reproduce

The AM-RP-002 research page links the calculator, four invented cases, counterexample tests and a contract guide. These run offline without a model, credentials or facility access. Previous collection-ledger and observation-evidence files remain unchanged.

The companion essay explains the interpretation limits. Next dependencies include authorized records, verified instruments and stock methods, reviewed tolerances, downstream evidence, a real review interface and accountable technical scrutiny. Policy simulation, field effectiveness, DGX experiments, peer review and scholarly indexing remain unestablished.

Proposed approach

Implemented
Offline reconciliation reference
Proposed
Evidence and operator review

Technical FAQ

Does a balanced account prove that material was recycled?
No. It establishes consistency among the declared quantities only. Physical measurements, source records and evidence of downstream processing or use require separate validation.
Can outgoing mass divided by incoming mass be called a recovery yield?
Not automatically. Opening inventory and material retained at closing can mix cohorts. State the attribution method and evidence before describing the yield of a particular input batch.
What does the released reference actually do?
It reconciles integer-gram quantities for one declared aggregate account, preserves missing values and signed discrepancies, and labels nonzero differences within an explicit tolerance. It runs no facility, AI model or collection policy.

Ready to architect operational leverage?

Available for fractional executive leadership, enterprise architecture consulting, and board-level technical advisory.

Initiate Dialogue ↗