GIS-Based Analysis of the Impacts of PM10 Concentration on Construction Accidents Using the Clustering Approach

Abstract Many studies have examined the health effects of particulate matter with a diameter of 10 μm or less ( PM 10 ) on construction workers, but its impact on construction accidents remains underexplored. This study investigates how PM 10 concentrations, influenced by geographic and seasonal variations, are statistically associated with construction accident risks. The goal is to enhance phased safety management strategies by classifying risk groups based on PM 10 levels. Using 13 years of big data, this study analyzes construction accidents and PM 10 concentrations across multiple regions and seasons. The methodology comprises four stages: data collection, risk assessment, cluster analysis, and statistical analysis. GIS–based techniques, including spatial autocorrelation and k -means clustering, were applied to identify accident risk groups based on PM 10 variations. Results revealed significant seasonal differences in accident risks. Spatial autocorrelation analysis, measured using Global Moran’s I to quantify the strength of regional clustering rather than accident magnitude, yielded values of 0.24 in spring, 0.19 in summer, 0.19 in autumn, and 0.46 in winter, indicating the strongest spatial concentration of similar risk levels during winter. K -means clustering classified accident risk into groups: spring (4), summer (4), autumn (3), and winter (4). Based on the average relative probability of accidents, a dimensionless metric representing accident likelihood normalized by PM 10 exposure duration, the highest accident risk was observed in spring, whereas summer exhibited the lowest risk. These results demonstrate that seasonal PM 10 conditions influence both the spatial concentration and relative magnitude of construction accident risk, supporting the need for region- and season-specific safety planning rather than uniform PM 10 management thresholds. This study introduces a novel risk classification framework incorporating regional and seasonal PM 10 variations. Unlike previous research focusing solely on health impacts, it provides a data-driven basis for targeted safety interventions, assisting policymakers and construction managers in prioritizing preventive measures and resource allocation under seasonally varying PM 10 -related risk conditions.

Authors

Institutions

Publication Details

Journal
ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems Part A Civil Engineering
Published
2026-11-13
DOI
https://doi.org/10.1061/ajrua6.rueng-1714
Primary Topic
Occupational Health and Safety Research
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

GIS-Based Analysis of the Impacts of PM10 Concentration on Construction Accidents Using the Clustering Approach

Hyeongjun Mun, Jaewook Jeong, Jaemin Jeong, Louis Kumi et al.
3 links
ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems Part A Civil Engineering
Occupational Health and Safety Research
article

GIS-Based Analysis of the Impacts of PM10 Concentration on Construction Accidents Using the Clustering Approach

Hyeongjun Mun, Jaewook Jeong, Jaemin Jeong, Louis Kumi, Taeseok Park
article en

Abstract

Abstract Many studies have examined the health effects of particulate matter with a diameter of 10 μm or less ( PM 10 ) on construction workers, but its impact on construction accidents remains underexplored. This study investigates how PM 10 concentrations, influenced by geographic and seasonal variations, are statistically associated with construction accident risks. The goal is to enhance phased safety management strategies by classifying risk groups based on PM 10 levels. Using 13 years of big data, this study analyzes construction accidents and PM 10 concentrations across multiple regions and seasons. The methodology comprises four stages: data collection, risk assessment, cluster analysis, and statistical analysis. GIS–based techniques, including spatial autocorrelation and k -means clustering, were applied to identify accident risk groups based on PM 10 variations. Results revealed significant seasonal differences in accident risks. Spatial autocorrelation analysis, measured using Global Moran’s I to quantify the strength of regional clustering rather than accident magnitude, yielded values of 0.24 in spring, 0.19 in summer, 0.19 in autumn, and 0.46 in winter, indicating the strongest spatial concentration of similar risk levels during winter. K -means clustering classified accident risk into groups: spring (4), summer (4), autumn (3), and winter (4). Based on the average relative probability of accidents, a dimensionless metric representing accident likelihood normalized by PM 10 exposure duration, the highest accident risk was observed in spring, whereas summer exhibited the lowest risk. These results demonstrate that seasonal PM 10 conditions influence both the spatial concentration and relative magnitude of construction accident risk, supporting the need for region- and season-specific safety planning rather than uniform PM 10 management thresholds. This study introduces a novel risk classification framework incorporating regional and seasonal PM 10 variations. Unlike previous research focusing solely on health impacts, it provides a data-driven basis for targeted safety interventions, assisting policymakers and construction managers in prioritizing preventive measures and resource allocation under seasonally varying PM 10 -related risk conditions.

ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems Part A Civil EngineeringVol. 13(1)
Seoul National University of Science and Technology (KR), University of Toronto (CA), Georgia Southern University (US)
Sustainable cities and communities, Good health and well-being
Occupational Health and Safety Research
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