ルーティングよりタイミング – 時間別系統炭素強度下での交換式バッテリーを用いた炭素配慮型トラック・ドローン配送
Code and data for: Timing over routing – carbon-aware truck–drone delivery with swappable batteries under hourly grid carbon intensity (原題)
Egemen HOPALI
🤖 gxceed AI 要約
日本語
電動ラストマイル配送の炭素排出が電力消費時刻に依存するかを検証した研究。並行トラック・ドローンの混合整数モデルで、バッテリー交換・充電開始時刻・ソーティ出発時刻を時間別系統炭素強度で価格付けし最適化する。年平均係数は46%のケースで日排出を10%超誤り、13系統中12で過小評価。炭素配慮型のタイミング調整は再ルーティングではなく充電・出発時刻の変更により平均7.8%（最大55%）排出を削減する。
English
This study tests whether the carbon footprint of electrified last-mile delivery depends on when electricity is drawn, and whether carbon-aware operation needs new routes or only re-timing. A mixed-integer truck–drone model prices each kWh at hourly grid carbon intensity and optimizes van tours, drone/van assignment, sortie launch times and battery charging. Annual-average factors misstate daily emissions by >10% in 46% of cases and understate year-round emissions in 12 of 13 grids; carbon-aware timing cuts realized emissions 7.8% on average (up to 55%), almost entirely via re-timing rather than re-routing.
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
日本では再エネ比率の時間変動と系統炭素強度の可視化が進む中、物流・小売のScope 3排出削減と電力調達の時間シフトは実務課題。SSBJのScope 3開示や省エネ法対応と接続しうる。
In the global GX context
Directly informs Scope 3 logistics accounting and hourly matching debates under CSRD/ISSB and 24/7 carbon-free energy procurement. Challenges annual-average emission factors used in corporate inventories and offers a re-timing lever for fleet electrification.
👥 読者別の含意
🔬研究者:時間別炭素強度を組み込んだ配送最適化と排出会計の誤差評価の手法を学べる。
🏢実務担当者:年平均係数の誤差と、充電・出発時刻シフトによる低コスト削減の可能性を物流計画に活用できる。
🏛政策担当者:排出係数の時間粒度の重要性と、電力・物流政策での時間別炭素シグナル活用を示唆。
📄 Abstract（原文）
This dataset contains the complete code, the processed open input data and the raw results behind the article "Timing over routing: Carbon-aware truck–drone delivery with swappable batteries under hourly grid carbon intensity". The study examines whether the carbon footprint of electrified last-mile delivery depends on the hour in which electricity is drawn, and whether carbon-aware operation requires new routes or only re-timing. A mixed-integer model for parallel truck–drone delivery decides the van tour, the drone-or-van assignment of customers, sortie launch times, the charging start of every swappable battery pack and the van's overnight charging, and prices each kilowatt-hour at the hourly carbon intensity of the grid. The model is solved exactly by a logic-based Benders decomposition (HiGHS 1.15.1). All inputs are open: benchmark delivery instances on the road networks of Buffalo and Seattle (multiple flying sidekicks benchmark, Murray & Raj, 2020), hourly lifecycle and direct carbon intensity for 13 grids in 2021 (CarbonCast, Maji et al., 2022), the 2026 UK greenhouse gas conversion factors (DESNZ) and a summary of the Electricity Maps signal coverage file. Folders: code/ (model, decomposition algorithm, experiment drivers, analysis and figure scripts in Python), data/ (processed instances and hourly intensity series), results/ (one row per instance-day and planning policy, about 15,000 instance-day runs), tables/ (stats.json holds every number quoted in the article, plus summary tables), figures/ (TIFF at 1000 dpi and PNG). The files reproduce the main findings: annual-average factors misstate the emissions of a delivery day by more than 10% in 46% of cases and understate year-round emissions in 12 of 13 grids, and carbon-aware timing lowers realized emissions by 7.8% on average (up to 55%), almost entirely by re-timing charging and sorties rather than by re-routing.
🔗 Provenance — このレコードを発見したソース
- openalex https://doi.org/10.17632/9g4gdh5n78first seen 2026-11-05 05:18:23
🔔 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。