Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Kundan Thota, Thorsten Schlachter, Veit Hagenmeyer

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.

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

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.
Paper Structure (25 sections, 4 equations, 6 figures, 6 tables, 1 algorithm)

This paper contains 25 sections, 4 equations, 6 figures, 6 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Metadata for Urban Energy Models
  4. Image-Based Building-Age Inference
  5. LLM-Driven Multi-Agent systems
  6. Proposed Method
  7. Multi‐Agent System for Data Collection
  8. Zensus agent
  9. OSM agent
  10. Monument agent
  11. Data orchestration, fusion and harmonization
  12. BuildingAgeCNN for Age Cohort Mapping
  13. Inference module
  14. Experimental Results
  15. Dataset coverage analysis
  16. ...and 10 more sections

Figures (6)

  • Figure 1: Outline: (1) Multiple data sources (databases, web pages, maps, and PDFs) are ingested into (2) LLM-driven agents to produce (3) geocoded age cohorts.
  • Figure 2: Overview of the proposed pipeline for building-age cohort mapping: A multi-agent system for building-age cohort mapping consists of three data‐collection agents (Zensus, OSM, Monument), then fed into a Data Orchestration, Fusion, and Harmonization that integrates and standardizes records into five age cohorts. The BuildingAgeCNN model (ConvNeXt + FPN + CoordConv + SE) is then trained on satellite imagery for supervised learning and inference.
  • Figure 3: Inference pipeline of the AgeCohort agent: address parsing, geocoding, satellite-tile retrieval, BuildingAgeCNN inference, and confidence-based flagging for manual review.
  • Figure 4: Representative roof types across age cohorts (left to right: pre-1919, 1919–1950, 1951–1978, 1979–2000, post-2000).
  • Figure 5: Data combined into six spatially contiguous folds, with each fold visualized as a region polygon.
  • ...and 1 more figures