Q-PROCツールキット：再現可能な実装と匿名化データセット（コスタリカ、サン・カルロス、2020-2025年）

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Provides a reproducible method and dataset for Scope 3 estimation from procurement data, with sensitivity analysis.

🏢実務担当者:Offers a practical toolkit for municipalities and companies to estimate Scope 3 emissions from procurement records without direct measurement.

🏛政策担当者:Demonstrates a low-cost approach for public sector GHG disclosure, informing policy on data requirements for procurement systems.

Q-PROC toolkit — reproducible implementation and anonymised dataset for recovering physical quantity from municipal public procurement records, applied to cantonal road asphalt paving in a Costa Rican municipality, 2020–2025. IPSASB Sustainability Reporting Standard SRS 1 requires annual GHG disclosure. For a local government, road asphalt is typically the most material capital good after waste, and falls under GHG Protocol Scope 3 Category 2 — which requires tonnes of mix placed , not currency spent. Municipalities rarely hold that figure. What they hold is a procurement record. This toolkit implements the seven rules (R1–R7) of the Q-PROC method: perimeter definition, material identification tolerant to catalogue misspelling, separation of placed work from ancillary services, assignment of physical magnitude, treatment of negative lines, thickness recovery and imputation, and conversion using externally sourced parameters. Result for the case: 1,258,475.0 m² · 129,105.8 t · 10,368.22 tCO2e over six years. Contents. recalculo.py (the seven rules), parametros.yaml (five parameters, each with its source), three anonymised CSV files (annual result, 418-line detail, nine sensitivity scenarios), README.md , LICENSE , CITATION.cff and SHA-256 checksums. Declared limitations. Imputed layer thickness moves the total by ±10.2%, against 1.76% for mix density — roughly six-fold; the highest-value improvement is therefore a mandatory thickness field in tender documents, not a better emission factor. The granular material density (1.6 t/m³) has no verified primary source and governs 19.1% of the result; it is flagged as pending in parametros.yaml . Haulage emissions correspond to granular material, not to the asphalt mix itself, whose transport is subsumed in the surfacing line price. Only 7.3% of the recovered mass was weighed directly by the entity; external validation is pending. Data protection. The underlying procurement record is not published, as it contains supplier identities, purchase-order numbers and unit prices. The derived tables allow full reproduction and audit of every figure reported without reconstructing it. Transferability. The method operates on line-level procurement data available in any OCDS-compliant system: SICOP (CR), SECOP II (CO), ComprasNet (BR), Mercado Público (CL), CompraNet (MX), SEACE (PE). Resumen en español. Implementación reproducible del método Q-PROC, que reconstruye la cantidad física de mezcla asfáltica colocada a partir del detalle de línea de órdenes de compra municipales, cuando la entidad no dispone de mediciones de obra. Incluye las siete reglas, los parámetros con su fuente, el conjunto de datos anonimizado y el análisis de sensibilidad de nueve escenarios. Los datos publicados no contienen proveedor, número de orden ni precio unitario.

• Zenodo https://zenodo.org/records/22168206first seen 2026-08-30 04:12:07 · last seen 2026-09-13 04:12:23
• openalex https://doi.org/10.5281/zenodo.22168205first seen 2026-09-01 05:15:23 · last seen 2026-09-01 05:15:25