Evidence-Grounded Multi-Agent Planning Support for Urban Carbon Governance via RAG

TL;DR

The paper addresses the challenge of integrating heterogeneous evidence for urban carbon governance and enabling reliable, traceable planning with LLMs. It introduces an evidence-grounded, four-agent planning system built on a standard text-based RAG pipeline (no GraphRAG) and a multi-source knowledge base to support accountability and cross-department collaboration. The evaluation comprises factual retrieval and planning-generation tasks, including a Ningbo, China case study, showing substantial improvements in grounding and coherent end-to-end reports, albeit with practical issues like data-boundary inconsistencies. The work demonstrates a practical, deployable approach for decision-support in urban carbon governance with clear pathways for real-world adoption and further refinement.

Abstract

Urban carbon governance requires planners to integrate heterogeneous evidence -- emission inventories, statistical yearbooks, policy texts, technical measures, and academic findings -- into actionable, cross-departmental plans. Large Language Models (LLMs) can assist planning workflows, yet their factual reliability and evidential traceability remain critical barriers in professional use. This paper presents an evidence-grounded multi-agent planning support system for urban carbon governance built upon standard text-based Retrieval-Augmented Generation (RAG) (without GraphRAG). We align the system with the typical planning workflow by decomposing tasks into four specialized agents: (i) evidence Q\&A for fact checking and compliance queries, (ii) emission status assessment for diagnostic analysis, (iii) planning recommendation for generating multi-sector governance pathways, and (iv) report integration for producing planning-style deliverables. We evaluate the system in two task families: factual retrieval and comprehensive planning generation. On factual retrieval tasks, introducing RAG increases the average score from below 6 to above 90, and dramatically improves key-field extraction (e.g., region and numeric values near 100\% detection). A real-city case study (Ningbo, China) demonstrates end-to-end report generation with strong relevance, coverage, and coherence in expert review, while also highlighting boundary inconsistencies across data sources as a practical limitation.