← 論文一覧に戻る

グリーン・アルゴリズム：リーンオペレーションと環境サステナビリティを橋渡しする人工知能

THE GREEN ALGORITHM: ARTIFICIAL INTELLIGENCE AS A BRIDGE BETWEEN LEAN OPERATIONS AND ENVIRONMENTAL SUSTAINABILITY (原題)

Hairin Nisa Meor Nordin, Zamri Miskan, Anis Shafiqah Mohd Saidun, Nur Qamarina Khairulanuar, Ashvanie Menon

International Journal of Innovation and Industrial Revolution📚 査読済 / ジャーナル2026-09-14#AI×ESGOrigin: Global経営インパクト: コスト削減対象セクター: cross_sector
DOI: 10.35631/ijirev.826018
原典: https://doi.org/10.35631/ijirev.826018
📄 PDF

🤖 gxceed AI 要約

日本語

RBVと動的ケイパビリティ理論に基づき、AI主導の業務統合が環境パフォーマンスに与える因果関係をUnilever・Amazon・Teslaの事例で検証。予測的ロードプーリングで物流CO₂64%削減、専用ハードで計算エネルギー29%削減、画像認識リサイクルで金属92%回収などの効果を示す。一方、AI運用自体の電力消費が便益を相殺しうる「AI-サステナビリティ・パラドックス」を指摘し、Scope 3報告の分断解消には業界横断のデータ基盤が必要と論じる。

English

Using RBV and dynamic capabilities theory, this study tests how AI-driven operational integration affects environmental performance through case studies of Unilever, Amazon, and Tesla. It reports 64% logistics CO₂ cuts via predictive load pooling, 29% computing energy savings from custom hardware, and 92% metal recovery through vision-assisted recycling. The authors flag an 'AI-Sustainability Paradox'—AI's own energy use may offset gains—and argue that breaking down Scope 3 data silos requires cross-sector, decentralized data systems.

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

📝 gxceed 編集解説 — Why this matters

日本のGX文脈において

AI活用による排出削減効果とAI自体の電力消費のトレードオフを扱う点は、SSBJのScope 3開示や有報での気候関連情報開示を進める日本企業にとって、AI導入の環境便益をどう測定・説明するかの示唆となる。

In the global GX context

The paper speaks directly to ISSB/CSRD-era disclosure debates: it links AI-enabled operational efficiency to measurable emissions outcomes while warning that AI's own energy footprint can undermine reported gains, and it calls for decentralized data architectures to overcome Scope 3 reporting silos—relevant to global disclosure infrastructure design.

👥 読者別の含意

🔬研究者:AI導入と環境パフォーマンスの因果経路をRBV/動的ケイパビリティで整理し、Jevons型パラドックスの実証枠組みとして参照できる。

🏢実務担当者:AIによる物流・製造・リサイクル最適化の定量効果と、AI電力消費を差し引いた正味便益の測定・開示設計に活用できる。

🏛政策担当者:AIの環境便益を過大評価しないよう、Scope 3報告のデータ分断解消とAI電力消費の開示枠組み整備を検討する材料になる。

📄 Abstract（原文）

Currently, industries around the world are under growing pressure to meet the United Nations Sustainable Development Goals (SDGs). Creating a reliable way to measure the environmental benefits of digital transformation has become essential for industries everywhere. This study builds a framework based on the Resource-Based View (RBV) and dynamic capabilities theory and aims to test the cause-and-effect link between AI-driven operational integration and environmental sustainability performance. Using a detailed study of leading companies—Unilever, Amazon, and Tesla—the research looks at how Resource Optimization plays a role in this connection. The results show that the independent factor has a strong effect on the dependent factor through several clear ways. For example, predictive load pooling helps cut logistics-related CO₂ emissions by 64%, supporting SDG 13. Also, custom hardware designs reduce computing energy use by 29%, matching SDG 9, while computer vision-assisted manufacturing recycling recovers 92% of metals, contributing to SDG 12. However, comparing cases reveals an important limit called the AI-Sustainability Paradox. This means the large energy needed to run the independent variable can cause a digital version of the Jevons Paradox, which might cancel out the environmental gains in the dependent variable. The study finds that while AI is essential to separate industrial growth from carbon emissions, its success relies on breaking down technology barriers between sectors and creating decentralized data systems to remove isolated Scope 3 reporting.

🔗 Provenance — このレコードを発見したソース

🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。

gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。