AI-Based Sustainable Supply Chain Management: Enhancing Transparency and Reducing Carbon Footprint

DOI

10.2991/978-94-6239-624-1_29How to use a DOI?

Keywords

Scope 3 Emissions; Sustainable Development Goals (SDGs); Natural Language Processing (NLP); Carbon Footprint; Machine Learning

Abstract

Achieving the Sustainable Development Goals (SDGs) poses a distinct challenge for large companies, especially concerning Goal 13 on climate change mitigation. One of the key challenges in meeting this goal is addressing Scope 3 emissions, which make up more than 90% of corporate emissions. However, dealing with these emissions is challenging due to the intricate web of upstream and downstream suppliers from which the emission data must be collected. This paper presents a novel approach using domain-adapted Natural Language Processing (NLP) foundation models to estimate Scope 3 emissions by automating transaction classification into commodity classes using financial transaction data. The approach is evaluated against text classification benchmarks, TF-IDF, Word2Vec, and zero-shot learning. Our results indicate that domain-adapted models significantly outperform classical techniques, as evidenced by the Word2Vec model achieving a higher F1 score of 72% compared to 69% for TF-IDF. Zero-shot classification based on commodity descriptions yielded better results than using titles, producing F1 scores from 40.1% to 43.7%. A fine-tuned Roberta-base model achieved the highest F1 score of 87.2%, improving to 87.19% with a lower learning rate. These results show the effectiveness of automating Scope 3 emissions estimation at an enterprise level, enabling proactive climate action in alignment with SDG 13 for reduced carbon emissions.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Kamana Parvej Mishu
AU  - Mohammad Tahmid Ahmed
AU  - Rahima Binta Bellal
AU  - Md. Abubakkar
AU  - Maisha Hossain Ahona
AU  - Ahmed Hassaan
AU  - Zeeshan Akbar
AU  - Efat Ara Haque
PY  - 2026
DA  - 2026/04/06
TI  - AI-Based Sustainable Supply Chain Management: Enhancing Transparency and Reducing Carbon Footprint
BT  - Proceedings of the International Conference on Sustainable Economics and Finance in the Digital Business Transformation (INCOSEF 2025)
PB  - Atlantis Press
SP  - 391
EP  - 402
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6239-624-1_29
DO  - 10.2991/978-94-6239-624-1_29
ID  - Mishu2026
ER  -