Explainable machine learning with multi-source data for community fire risk prediction

Fires pose a significant threat to citizens’ lives, property, and overall urban safety. While research has extensively focused on forest fire risk assessment and prediction, comparatively little attention has been given to evaluating fire danger in large metropolitan areas. This study integrates data from Yuncheng City, China – including building characteristics, road networks, population distribution, fire occurrence records, and meteorological parameters to facilitate a geographically informed analysis. Various machine learning models were trained to assess fire risk throughout the city. The findings demonstrate that the XGBoost model exhibits high predictive accuracy and strong spatial correlation, rendering it suitable for geographic analysis. This research identifies potential fire risks within urban areas, offering a scientific basis for the strategic deployment of firefighting resources and informed urban planning to mitigate losses of life and property.

Authors

Institutions

Publication Details

Journal
Proceedings of the Institution of Civil Engineers - Engineering Sustainability
Published
2026-11-19
DOI
https://doi.org/10.1680/jensu.26.00118
Primary Topic
Safety and Risk Management
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Explainable machine learning with multi-source data for community fire risk prediction

Ding Wang, Shan Xu, Yue Tian, Mengyu Chen et al.
4 links
Proceedings of the Institution of Civil Engineers - Engineering Sustainability
Safety and Risk Management
article

Explainable machine learning with multi-source data for community fire risk prediction

Ding Wang, Shan Xu, Yue Tian, Mengyu Chen, Feng Yue, Huiyu Cheng, Hengshao Wan
article en

Abstract

Fires pose a significant threat to citizens’ lives, property, and overall urban safety. While research has extensively focused on forest fire risk assessment and prediction, comparatively little attention has been given to evaluating fire danger in large metropolitan areas. This study integrates data from Yuncheng City, China – including building characteristics, road networks, population distribution, fire occurrence records, and meteorological parameters to facilitate a geographically informed analysis. Various machine learning models were trained to assess fire risk throughout the city. The findings demonstrate that the XGBoost model exhibits high predictive accuracy and strong spatial correlation, rendering it suitable for geographic analysis. This research identifies potential fire risks within urban areas, offering a scientific basis for the strategic deployment of firefighting resources and informed urban planning to mitigate losses of life and property.

Proceedings of the Institution of Civil Engineers - Engineering Sustainability
Yanshan University (CN)
National Natural Science Foundation of China
Sustainable cities and communities
Safety and Risk Management
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon