A Multi-Agent System for Building-Age Cohort Mapping to Support Urban Energy Planning

2026-03-18

AI Summary

This study addresses the challenge of missing and inconsistent building age data, which hinders sustainable district heating planning. To overcome this limitation, the authors propose the first framework that integrates multi-agent large language models with deep learning for inferring building construction years. The approach employs a multi-agent system to harmonize census records, OpenStreetMap data, and heritage registries, followed by geocoding and deduplication. A novel BuildingAgeCNN model is then trained, featuring a ConvNeXt backbone enhanced with FPN, CoordConv, and SE modules, alongside a confidence calibration mechanism to support low-risk decision-making. Under spatial cross-validation, the model achieves an overall accuracy of 90.69% and a macro F1-score of 67.25%, demonstrating its effectiveness in enabling optimized district heating and low-carbon energy deployment.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Modeling other Agents
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Determining the age distribution of the urban building stock is crucial for sustainable municipal heat planning and upgrade prioritization. However, existing approaches often rely on datasets gathered via sensors or remote sensing techniques, leaving inconsistencies and gaps in data. We present a multi-agent LLM system comprising three key agents, the Zensus agent, the OSM agent, and the Monument agent, that fuse data from heterogeneous sources. A data orchestrator and harmonizer geocodes and deduplicates building imprints. Using this fused ground truth, we introduce BuildingAgeCNN, a satellite-only classifier based on a ConvNeXt backbone augmented with a Feature Pyramid Network (FPN), CoordConv spatial channels, and Squeeze-and-Excitation (SE) blocks. Under spatial cross validation, BuildingAgeCNN attains an overall accuracy of 90.69% but a modest macro-F1 of 67.25%, reflecting strong class imbalance and persistent confusions between adjacent historical cohorts. To mitigate risk for planning applications, the address-to prediction pipeline includes calibrated confidence estimates and flags low-confidence cases for manual review. This multi-agent LLM system not only assists in gathering structured data but also helps energy demand planners optimize district-heating networks and target low-carbon sustainable energy systems.

Problem

• building age estimation
• urban energy planning
• data fusion
• building stock mapping
• sustainable heating

Innovation

• multi-agent LLM system
• BuildingAgeCNN
• ConvNeXt with FPN
• heterogeneous data fusion
• calibrated confidence estimation