Abstract:

To identify the critical causes and paths of bridge girder-erecting machine accidents in construction engineering, a risk factor analysis method based on complex network combined with N-K model was proposed. Firstly, the cases database of bridge erecting machine accidents between 2000-2022 was established, and the accident causes of bridge girder-erecting machine were classified into four first-level risk factors and twenty-four second-level risk factors by analyzing the cause of the accidents. Then, a complex network model with major accident types and the second-level risk factors as nodes and their correlations as edges was established, and the closeness centrality and betweenness centrality of risk factor nodes were calculated. Finally, the N-K model was developed for the first-level risk factor coupling analysis, and the centrality of the risk factor nodes was optimized by using the coupled interactive information. Through the accessibility analysis of risk factors and global main path analysis, 7 key causes and 7 key links of bridge machine accidents were obtained respectively. The results show that the weak management of construction unit and the insufficient safety production inspection have the high value of closeness centrality and betweenness centrality. The biggest risk is from the path of duty dereliction of supervision unit→the weak management of construction unit→the insufficient safety production critical. So, all these key causes and paths can be taken as the focus of safety control during the construction of bridge girder-erecting machine. The key cause of bridge girder-erecting machine construction safety obtained by complex network integrating with N-K model is consistent with the conclusion of accident investigation, and the risk identification of girder-erecting machine construction safety accident is accurately realized.

Key words: N-K model, complex network, bridge girder-erecting machine, construction safety risk factors, critical cause, critical path