フードシステム炭素データレイヤー：自動ライフサイクル評価デジタルツインの都市デジタルツインへの統合

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

🔬研究者:AI自動LCAと都市代謝会計の境界整合・不確実性伝播を扱う新規研究領域を提示する。

🏢実務担当者:食品調達のScope 3算定や自治体炭素データ連携の設計に、境界整合と不確実性管理の枠組みを提供する。

🏛政策担当者:都市低炭素計画にフードシステム排出を組み込む際のデータ基盤とガバナンス条件を示唆する。

Urban digital twins for climate-responsive planning routinely model energy consumption, mobility emissions and thermal comfort at city scale, yet food-system greenhouse gas emissions, between roughly one-quarter and one-third of global anthropogenic emissions depending on accounting boundary, remain absent from urban carbon data infrastructure and municipal low-carbon planning. A maturing class of AI-driven automated life cycle assessment platforms now estimates product-level carbon footprints for tens of thousands of food products at supply-chain resolution, several of which providers describe as digital twins updating as primary supplier data changes. A documented scan and individual screening of the intersecting literature identifies adjacent contributions, including an urban digital twin framework for consumption-based neighbourhood carbon footprints, but no framework that formalizes automated food-system life cycle assessment twins as a structured, uncertainty-aware carbon data layer with an explicit boundary crosswalk and uncertainty propagation. This paper proposes the Food-System Carbon Data Layer, a conceptual architecture positioning these twins as interoperable inputs to human-centered urban digital twin platforms. We advance a two-part hypothesis: that the integration is analytically feasible on publicly available life cycle inventory data, and that its operational viability is governance-dependent under emerging regulatory instruments. The framework addresses four challenges: boundary alignment between cradle-to-gate product carbon footprint models and urban metabolism accounting; uncertainty propagation from inventory data-gap proxies and machine learning classifiers through to city-scale dashboards, preserving producer-level variance rather than collapsing it into ordinal grades; semantic harmonization across the vocabularies of procurement records, supplier bills of materials and inventory databases; and human-centered representation for planners, procurement officers and public health authorities with heterogeneous data literacy. Four worked urban use cases on publicly available inventory data illustrate the framework, and we specify the interoperability conditions for viable integration. The use cases establish analytical feasibility, meaning the architecture is specifiable and computable on open data; they are not an empirical validation against a deployed urban digital twin, which we identify as the next required step. H1 is stated with four checkable success criteria, two carrying declared numerical thresholds, and H2 with the panel that would test it.

• openalex https://doi.org/10.3389/fbuil.2026.1953330first seen 2026-09-30 05:02:26