Cause analysis of construction collapse accidents using association rule mining

Abstract

PurposeThe construction collapse is one of the most serious accidents since it has several attributes (e.g. accident type and consequence) and its occurrence involves various kinds of causal factors (e.g. human factors). The impact of causal factors on construction collapse accidents and the interrelationships among causal factors remain poorly explored. Thus, the purpose of this paper is to use association rule mining (ARM) for cause analysis of construction collapse accidents.Design/methodology/approachAn accident analytic framework is developed to determine the accident attributes and causal factors, and then ARM is introduced as the method for data mining. The data are from 620 historical accident records on government websites of China from 2010 to 2020. Through the generated association rules, the impact of causal factors and the interrelationships among causal factors are explored.FindingsCollapse accident is easily caused by human factors, material and machine condition and management factors. Furthermore, the results show a close interrelationship between many causal factors and construction scheme and organization. The earthwork collapse is greatly related to environmental condition and the scaffolding collapse is greatly related to material and machine condition.Practical implicationsThis study found relevant knowledge about the key causes for different types of construction collapses. Besides, several suggestions are further provided for construction units to prevent construction collapse accidents.Originality/valueThis study uses data mining methods to extract knowledge about the causes of collapse accidents. The impact of causal factors on various types of construction collapse accidents and the interrelationships among causal factors are explained from historical accident data.

Interaction mechanisms of interface management risks in complex systems of high-speed rail construction projects: an association rule mining-based modeling framework

“…At the same time, confidence represents the conditional probability of Y occurring in the presence of X . Association rules that satisfy both the minimum threshold for support and confidence and have a lift value greater than one are commonly known as effective strong association rules (Shao et al ., 2022). In this study, we utilized the Apriori algorithm to extract rules related to IMRs in high-speed rail projects.…”

Establishment of Safety Management Measures for Major Construction Workers through the Association Rule Mining Analysis of the Data on Construction Accidents in Korea

“…Many causal factors, such as construction site environment, supervisory staffing, and inspections, are related to safety management planning and execution. Their importance has been emphasized in previous studies [37,38]. Based on the ARM and network graph analyses, we summarized the safety management directions for each major accident type.…”

Risk Analysis of Laboratory Fire Accidents in Chinese Universities by Combining Association Rule Learning and Fuzzy Bayesian Networks

“…Cabello et al [29] used ARs to identify the main influencing factors leading to accidents in the construction phases. Shao et al [33] employed ARs to identify the causes of building collapse accidents. Those studies show that the application of ARs has the capability to identify and describe probabilistic relationships among associated factors, and further, it helps to develop BN models.…”