Risk Factor Analysis of Bridge Operation and Maintenance Safety Accidents Based on Text Mining

To effectively prevent bridge operation and maintenance (O&M) accidents, a hybrid method integrating text mining with Bayesian network theory was adopted to systematically analyze the causal factors of accidents affecting bridge O&M safety and their interrelationships. First, text mining techniques were applied to preprocess data from 226 Chinese bridge O&M accident cases (1999–2025). Using the BERTopic topic modeling approach, 20 safety risk topics were identified, and 20 corresponding causal factors were extracted based on key feature words under these topics. Drawing on the Human‐Machine‐Environment‐Management safety management theory, the safety risk causal factors were classified into four risk categories, namely, bridge self‐factors, O&M environmental factors, human factors, and management defects. The Apriori algorithm was employed to analyze the interrelationships among risk factors, from which a total of 42 strong association rules were extracted; these rules were then taken as the basis for constructing a Bayesian network model. Through Bayesian network reasoning and sensitivity analysis, the key causal factors, sensitive factors, and the maximum‐probability causal chain of bridge O&M safety accidents were clarified. Analysis of the key causal chain revealed that the chain with the highest probability of leading to bridge O&M safety accidents is as follows: inadequate industrial supervision→substandard construction quality→structural construction defects→insufficient structural bearing capacity→bridge accidents. Sensitivity analysis identified vessel collision and floating debris collision as the key sensitive factors influencing bridge O&M safety accidents, whereas insufficient structural bearing capacity and other relevant factors were defined as moderately sensitive factors, thus providing a robust basis for the targeted prevention and control of such accidents. This method avoids the interference of experts' subjective factors, conducts an in‐depth analysis of bridge O&M safety accidents based on objective data, and provides theoretical support and methodological guidance for formulating more scientific accident prevention measures, which endows it with important practical significance in bridge engineering practice.

Authors

Institutions

Publication Details

Journal
Safety Science and Technology
Published
2026-11-18
DOI
https://doi.org/10.1002/sst3.70060
Primary Topic
Construction Engineering and Safety
Type
article
Field-Weighted Citation Impact
0.00

Funders

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

Risk Factor Analysis of Bridge Operation and Maintenance Safety Accidents Based on Text Mining

Zhiguo Shao, Xingting Huang
3 links
Safety Science and Technology
Construction Engineering and Safety
article

Risk Factor Analysis of Bridge Operation and Maintenance Safety Accidents Based on Text Mining

Zhiguo Shao, Xingting Huang
article en

Abstract

To effectively prevent bridge operation and maintenance (O&M) accidents, a hybrid method integrating text mining with Bayesian network theory was adopted to systematically analyze the causal factors of accidents affecting bridge O&M safety and their interrelationships. First, text mining techniques were applied to preprocess data from 226 Chinese bridge O&M accident cases (1999–2025). Using the BERTopic topic modeling approach, 20 safety risk topics were identified, and 20 corresponding causal factors were extracted based on key feature words under these topics. Drawing on the Human‐Machine‐Environment‐Management safety management theory, the safety risk causal factors were classified into four risk categories, namely, bridge self‐factors, O&M environmental factors, human factors, and management defects. The Apriori algorithm was employed to analyze the interrelationships among risk factors, from which a total of 42 strong association rules were extracted; these rules were then taken as the basis for constructing a Bayesian network model. Through Bayesian network reasoning and sensitivity analysis, the key causal factors, sensitive factors, and the maximum‐probability causal chain of bridge O&M safety accidents were clarified. Analysis of the key causal chain revealed that the chain with the highest probability of leading to bridge O&M safety accidents is as follows: inadequate industrial supervision→substandard construction quality→structural construction defects→insufficient structural bearing capacity→bridge accidents. Sensitivity analysis identified vessel collision and floating debris collision as the key sensitive factors influencing bridge O&M safety accidents, whereas insufficient structural bearing capacity and other relevant factors were defined as moderately sensitive factors, thus providing a robust basis for the targeted prevention and control of such accidents. This method avoids the interference of experts' subjective factors, conducts an in‐depth analysis of bridge O&M safety accidents based on objective data, and provides theoretical support and methodological guidance for formulating more scientific accident prevention measures, which endows it with important practical significance in bridge engineering practice.

Safety Science and Technology
Qingdao University of Science and Technology (CN), Qingdao University of Technology (CN)
National Natural Science Foundation of China, Natural Science Foundation of Shandong Province
Industry, innovation and infrastructure
Construction Engineering and Safety
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