Automated IFC Generation and Machine Learning-Based λ-Correction for Embodied Carbon Estimation of Buildings

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

🔬研究者:This work provides a validated ML correction for simplified IFC-based embodied carbon estimates, advancing BIM-LCA integration and data-driven carbon accounting research.

🏢実務担当者:The framework allows embodied carbon estimation of existing buildings with minimal inputs, enabling cost-effective portfolio screening and retrofit planning for carbon disclosure.

🏛政策担当者:The methodology can inform standardized simplified embodied carbon estimation protocols in building codes and carbon footprinting guidelines, especially for existing building stock.

本研究は、Industry Foundation Classes（IFC）モデルの自動生成とX補正技術を統合することで、建物の体積炭素（embodied carbon）を推定する実用的な枠組みを提案する。総床面積、階数、階高、構造タイプ、竣工年、建物用途という6つの最小限の建物属性を用いて、簡略化されたIFCモデルを自動生成した。材料厚さ、密度、上乗せ率のプリセット値を適用し、Environmental Product Declaration（EPD）データをリンクして基準排出量を算出した。しかし、基準となるIFC結果は実際の体積炭素のわずか5〜20%しか占めず、系統的な過小評価が確認された。この限界に対処するため、平均Xを用いたグローバルスケーリング、K近傍法を用いたコホートベース補正、Light Gradient Boosting Machine（LightGBM）を用いた機械学習回帰の3つの補正方法を評価した。トレーニング用260棟、テスト用44棟を含む304棟のデータセットを用いて検証を行った。結果は、スケーリングとコホートアプローチは限られた精度しか提供せず、機械学習モデルが最良の性能（R²=0.803、MAPE=23.3%）を達成したことを示した。これらの知見は、最小限の入力であっても、設計文書を欠く既存建物に対して信頼性の高い体積炭素推定が可能であることを実証している。提案された枠組みはBIM-LCA統合を支援し、カーボンニュートラルな建物設計と改修のためのデータ駆動型戦略に貢献する。

AI 翻訳（deepseek-v4-flash）。 正確を期す場合は下の原文を参照してください。

This study proposes a practical framework for estimating embodied carbon in buildings by integrating automated generation of Industry Foundation Classes (IFC) models with X-correction techniques. Six minimal building attributes—gross floor area, number of floors, floor height, structural type, year of completion, and building use—were used to automatically generate simplified IFC models. Preset values for material thickness, density, and surcharge rates were applied, and Environmental Product Declaration (EPD) data were linked to calculate baseline emissions. The baseline IFC results, however, accounted for only 5-20% of actual embodied carbon, confirming systematic underestimation. To address this limitation, three correction methods were evaluated: global scaling with an average X, cohort-based correction using K-nearest neighbors, and machine learning regression with Light Gradient Boosting Machine (LightGBM). A dataset of 304 buildings, including 260 for training and 44 for testing, was used for validation. Results showed that scaling and cohort approaches provided limited accuracy, while the machine learning model achieved the best performance (R 2 =0.803, MAPE=23.3%). These findings demonstrate that even with minimal inputs, reliable embodied carbon estimation is feasible for existing buildings lacking design documents. The proposed framework supports BIM-LCA integration and contributes to data-driven strategies for carbon-neutral building design and retrofitting.

• semanticscholar https://www.e3s-conferences.org/articles/e3sconf/pdf/2026/34/e3sconf_iaqvec2026_11008.pdffirst seen 2026-06-29 06:38:12 · last seen 2026-08-09 05:49:41