What question did this study set out to answer?

The aim is to develop deep neural network models for predicting fire risks on construction sites based on weather data.

April 27, 2026Open Access

A Data-Driven Framework for Proactive Construction Safety: Assessing Meteorological Hazards with Optimized Deep Neural Networks

Key Points

The aim is to develop deep neural network models for predicting fire risks on construction sites based on weather data.
Analyzed historical fire data and automated weather system data from 2010 to 2023 in South Korea.
Identified key predictive variables including minimum temperature and maximum wind speed, and developed a fire risk index (Risk 5).
Conducted an exhaustive search for optimal DNN architectures to evaluate prediction accuracy and efficiency.
Achieved high performance with an optimal model structure (32-16-16), yielding an R² value of 0.917.
XAI analysis confirmed the model's predictions were valid, revealing how key variables interact with fire risk.
Successfully visualized fire occurrence patterns using regional meteorological data.

Abstract

Construction sites face high fire risks significantly influenced by dynamic meteorological conditions. This study develops and validates deep neural network (DNN) models for the quantitative prediction of construction fire risk using weather variables. We analyzed fire history and automated weather system (AWS) data from 2010 to 2023 in South Korea, identifying minimum temperature, effective humidity, maximum wind speed, atmospheric pressure, and month as key predictors. A novel fire risk index (Risk 5) was developed using a weighted combinatorial probability approach based on actual fire frequency and selected as the optimal dependent variable. An exhaustive search of DNN architectures (1-3 layers; 8-64 nodes) was conducted. Evaluating prediction accuracy (R²), complexity (AIC/BIC), and efficiency (pareto frontier), the optimal model (32-16-16 structure) was selected. This model achieved high performance (R²=0.917). Explainable AI (XAI) analysis (SHAP, PDP) validated the model's scientific basis by confirming the influence of key variables and their interactions. Furthermore, geospatial visualization using regional meteorological data successfully reproduced actual seasonal and regional fire occurrence patterns. This study provides a robust, data-driven tool for proactive construction site fire safety management.

Demander à l'IA

Bookmark

View Full Paper