A multi-model approach to construction site safety: Fault trees, Bayesian networks, and ontology reasoning

Donghui Shi; Shuling Gan; Jozef Zurada; Jian Guan; Feilong Wang; Paweł Weichbroth

doi:10.1016/j.eswa.2025.127817

A multi-model approach to construction site safety: Fault trees, Bayesian networks, and ontology reasoning

Abstrakt

With the rapid expansion of the construction industry, accidents on construction sites have been increasingly common. Among these, collapses of earthwork foundation pits are particularly significant due to the sheer weight of the collapsed materials and the vast affected areas, resulting in substantial casualties and economic losses. This study aims to determine the causes of these construction safety and collapse incidents and understand their relationships to enable effective supervision and prevention during construction. Initially, the LDA model is used to categorize historical construction safety accident reports, and text mining is applied utilizing R language platform and the TF-IDF measure to extract keywords related to accident causative factors from historical accident reports. Following this, risk factors are evaluated to find the basic, intermediate, and top events of the accident, constructing a fault tree of casualties from earthwork foundation pit collapse (EFPC) accidents and analyzing the structural significance of risk factors. The fault tree is converted into a Bayesian network through image and numerical mapping, allowing the analysis of node sensitivity and the prediction of top event probability for informed construction accident prediction and prevention. Lastly, the study constructs an ontology knowledge base and knowledge graph in the realm of building safety, and establishes an ontology reasoning model using the Pellet reasoning machine and SWRL reasoning rules. The novel approach in this study involves integrating multiple advanced methodologies and provides a comprehensive framework and knowledge foundation for enhancing construction accident prediction and prevention as well as understanding and mitigating the risk-causing factors of earthwork foundation pit collapses in the construction industry.

Cytowania

0

CrossRef
0

Web of Science
0

Scopus

Autorzy (6)

Cytuj jako

Pełna treść

pełna treść publikacji nie jest dostępna w portalu

pełna treść artykułu zobacz w serwisie zewnętrznym otwiera się w nowej karcie

Słowa kluczowe

Informacje szczegółowe

Kategoria:

Publikacja w czasopiśmie

Typ:

artykuły w czasopismach

Opublikowano w:

EXPERT SYSTEMS WITH APPLICATIONS nr 288,
ISSN: 0957-4174

Język:

angielski

Rok wydania:

2025

Opis bibliograficzny:

Shi D., Gan S., Zurada J., Guan J., Wang F., Weichbroth P.: A multi-model approach to construction site safety: Fault trees, Bayesian networks, and ontology reasoning// EXPERT SYSTEMS WITH APPLICATIONS -, (2025), s.127817-

DOI: