GX & Decarbonization Research

Yu Qiao, Xinwen Deng, Xiaoying Huang +7

This study proposes a UAV-UGV collaborative delivery framework for urban last-mile logistics, developing a multi-objective optimization model considering efficiency, safety, and low-carbon performance. A visual simulation platform using HTM…

K. Zhou, Y. H. Cao, H. Yang

Using data from Shanghai and Shenzhen A-share listed companies (2010-2023), this study employs a two-way fixed-effects model to analyze the impact of venture capital (VC) on corporate carbon performance. VC improves carbon performance by al…

Qiusheng Gao, Meichen Yao, Liang Duan

This paper proposes a new paradigm for low-carbon retrofitting of existing wastewater treatment plants by coupling AI fine-tuned operation with dynamic in-situ retrofitting. AI optimizes operations to reduce energy consumption and greenhous…

Fahrettin Pala, Emine Kaya, Esra Nur Akpınar +3

Using panel data from 50 countries (2002-2019), this study examines how AI, renewable energy technology innovation (RETI), and regulatory quality (RQ) affect low-carbon energy transition. AI promotes transition across all income groups, whi…

Xiaojie Lin, Jian Li, Rui Jing +4

This paper reviews the role of AI in transforming energy systems for carbon neutrality. It highlights how data-driven prediction, physics-informed learning, reinforcement learning, digital twins, and generative models enhance renewable fore…

Jiachen Shen, Jian Shi, Hui Zhong +3

This paper applies mechanism-guided graph learning to evaluate carbon border policies (e.g., CBAM) in interconnected European electricity markets, integrating AI methods with climate policy assessment to quantify cross-border policy spillov…

Kollia M

This Perspective proposes the ZERO framework to shift AI data-centre sustainability evaluation upstream, before resource consumption. It combines hard admissibility constraints, workload characterisation, local resource-state assessment, fe…

Zhang Q, Sheng S, Liu T

A two-stage lexicographic framework for geo-distributed AI inference that preserves service while optimizing carbon and water. Trace-driven analysis of 26,392 requests reveals a 21-point carbon-water frontier, quantifying trade-offs. Provid…

Yao, Huaying

This replication package audits corporate climate-target claims against mandatory verified emissions from the EU Transaction Log and US EPA GHGRP. Using only public data, it parses 1,779 claims into six adjudication tiers and documents sens…

Yutong Shen, Wang Yang, Yue Shen

CogDeBias is a multilingual framework integrating LLMs and ML to automatically detect and mitigate six cognitive biases (e.g., confirmation bias, sunk cost fallacy) in corporate annual reports. A bilingual corpus of 1,200 reports (600 Engli…

Antans Sauhats, Diāna Žalostība, Roman Petrichenko +4

This paper proposes an AI-based surrogate modeling framework using artificial neural networks (ANNs) to accelerate long-term power system transition planning. By combining stochastic scenario generation with detailed simulations and ANN sur…

Ma Y.

This study uses interpretable machine learning to identify greenwashing in heavily polluting enterprises and uncover its multidimensional driving mechanisms. By analyzing corporate ESG disclosures and behavioral data, it builds a model that…

Usama Aslam, Vikram Kumar, Muhammad Ahsan Niazi +1

This paper proposes Graph-X, a unified graph-structured deep learning and stochastic optimization framework for day-ahead electricity price forecasting and risk-aware VPP bidding. It models market-clearing behavior by converting bids into p…

Yuhao Gu, Han Lai

This paper explores how FinTech (blockchain, AI, big data) enhances green finance efficiency through environmental transparency, risk mitigation, and green innovation. Empirical results show blockchain improves disclosure authenticity, AI r…

Andreas Poyias, Panayiotis Mourtopallas, Diamanto Platanou +3

This study develops a digital framework for optimizing green hydrogen integration in food industry steam production, which accounts for up to 57% of energy use. LightGBM models trained on PV and weather data achieve high forecasting accurac…

Firat Bolat, Hande Demi̇rel

This study integrates Sentinel-5P TROPOMI satellite observations with AIS data to monitor ship emissions in the Istanbul Strait, estimating annual CO2 of 213,678 t, NOx of 5,970 t, and SOx of 686 t. A correlation of 0.76 between AIS-derived…

Kjartan Sigurðsson

This chapter analyzes AI's transformative role in sustainability reporting, particularly ESG, arguing it represents institutional change rather than mere technological improvement. It examines how blockchain, generative AI, and real-time mo…

Thing-Yuan Chang, Chien-Chih Chen, Chin-Hsien Hsu +2

The food supply chain emits 13.7 billion tons of CO2e annually, yet accurate Product Carbon Footprint (PCF) estimation is inaccessible to many manufacturers. This study proposes a machine learning framework combining Gaussian Mixture Model …

Paul Kron

This paper estimates the carbon cost of a large language model (LLM) query using life-cycle analysis and greenhouse gas emissions, calculating an ultimate cost of approximately $0.4 per query (about 10 gCO2eq/query) for future generations. …

Dara Ghahremani

This repository provides a machine-readable corpus of annual reports from European listed banks (2013-2023), along with scripts and an ESG dictionary for text analysis. It supports reproducible ESG disclosure research, including term-densit…