Agentic AI Sustainability Assessment for Supply Chain Document Insights

TL;DR

The paper tackles the challenge of making document-intensive supply chain automation sustainable by introducing an agentic AI framework that operates within human-in-the-loop governance. It compares fully manual, AI-assisted HITL, and advanced multi-agent agentic workflows, demonstrating substantial eco-efficiency gains (up to 70–90% reduction in energy, 73–97% in CO$_2$, and 89–98% in water) while maintaining data quality. A replicable use case using Gemini 2.5 Flash with extended thinking validates the methodology and provides complete token accounting and energy metrics, published in a public GitHub repository. The work integrates sustainability into AI governance through an ESG-oriented framework and discusses strategic, resilience, and security considerations for scalable, responsible AI deployment in global supply chains.

Abstract

This paper presents a comprehensive sustainability assessment framework for document intelligence within supply chain operations, centered on agentic artificial intelligence (AI). We address the dual objective of improving automation efficiency while providing measurable environmental performance in document-intensive workflows. The research compares three scenarios: fully manual (human-only), AI-assisted (human-in-the-loop, HITL), and an advanced multi-agent agentic AI workflow leveraging parsers and verifiers. Empirical results show that AI-assisted HITL and agentic AI scenarios achieve reductions of up to 70-90% in energy consumption, 90-97% in carbon dioxide emissions, and 89-98% in water usage compared to manual processes. Notably, full agentic configurations, combining advanced reasoning (thinking mode) and multi-agent validation, achieve substantial sustainability gains over human-only approaches, even when resource usage increases slightly versus simpler AI-assisted solutions. The framework integrates performance, energy, and emission indicators into a unified ESG-oriented methodology for assessing and governing AI-enabled supply chain solutions. The paper includes a complete replicability use case demonstrating the methodology's application to real-world document extraction tasks.

Figures (4)

• Figure 1: Comparative sustainability metrics between AI-assisted (HITL) and human-only scenarios, showing maximum values for energy consumption, water usage, CO$_2$ emissions, and required human operators. The chart illustrates the dramatic reduction achieved through AI-assisted document processing at an operational scale of 5,000 documents per day.
• Figure 2: Comparative sustainability metrics between AI-assisted (HITL) and human-only scenarios. The bar chart illustrates the dramatic reduction in energy consumption (blue), water usage (red), CO$_2$ emissions (purple), and required human resources (green) achieved through AI-assisted document processing at an operational scale of 5,000 documents per day.
• Figure 3: Multi-agent workflow architecture illustrating the sequential processing pipeline: Parser Agent converts documents to markdown, Generative AI Agent processes content, Validator Agent ensures accuracy, and Human-in-the-Loop (HITL) performs final review and approval.
• Figure 4: