Coupling analysis of crane accident risks factors based on Bayesian network and the N-K model

Abstract

A crane is one of the most widely used pieces of equipment on construction sites, but it is accompanied by a high accident rate. Crane accident risk analysis is very important for good lifting accident prevention and control and ensuring the safety of lifting operations. Therefore, we should attach great importance to the interaction between different types of risk factors. This paper proposes a coupling quantitative method of human, machine, management, and environmental risk factors for crane accidents based on Bayesian networks (BN) and the N-K model. Firstly, text mining technology and fault tree analysis are used to analyze the causes of crane accidents and classify the risk factors. Secondly, the types of risk coupling caused by human, machine, management, and environmental risk factors are defined. Thirdly, based on the crane accident risk analysis and its N-K model, the BN model is established. Fourthly, the parameters of the risk coupling nodes in the developed BN are determined by the calculation results of the N-K model. Finally, the risk coupling type with a high coupling value and the first-level node and second-level node the failure probability caused is analyzed by posterior probability and sensitivity analysis. The results find that man risk factors and management risk factors have a significant impact on crane accidents, and should be strengthened; the coupling effect of multiple risk factors has a great influence on the probability of crane accidents, which should be paid much attention to.

Crane Operation Hazard Evaluation Based on Z-Number and CPT

“…combined fault tree analysis, fuzzy set theory, Bayesian networks, and Markov chains. Wu et al [15] presented a quantitative approach for crane accidents using Bayesian networks and an N-K model. Zheng et al [16] established a fault diagnosis system for a crane spreader based on a fault tree and a Bayesian network, which was based on 13 years of historical fault data collected from their online service.…”