Building construction involves complex interactions among personnel, machinery, and environmental factors, which can lead to risk incubation and unpredictable outcomes. However, most existing studies on safety risk coupling either rely on qualitative approaches or adopt single-method quantitative tools, which are limited in capturing the multidimensional and dynamic interactions of risks, especially in super-high-rise projects. This paper introduces a quantitative safety risk coupling assessment method based on complex network theory to enhance systematic safety risk management. The method begins by identifying safety risk events and factors using accident data. A safety risk evolution model is then developed, incorporating fault tree analysis (FTA) and accident chain (AC) analysis. The model creates a complex network of coupled risk events and factors, weighted through risk entropy (RE) analysis. Subsequently, the method assesses both occurrence-type coupling of factors and deterioration-type coupling of events. Occurrence-type coupling is evaluated using a multi-path evolution analysis with a depth-first search (DFS) algorithm, and deterioration-type coupling is examined through accident chain transfer analysis. Finally, quantitative risk coupling assessment results can be formed. This method was applied to super-high-rise building construction cases. The primary safety risk events with high frequency and significant losses were identified based on an analysis of 1,240 cases, including incidents such as tower crane collapses, falls of construction lift cages, collapses of attached lift scaffolds, falls from heights, and workers being struck by objects, among others, together with risk factors. A safety risk evolution model was developed, and the characteristics of the complex network were analyzed. In a case study of a project in Shanghai, China, the results of the quantitative risk assessment were compared with those of traditional risk level assessments. The findings demonstrate that the risk levels derived from the quantitative method align more closely with the actual perceptions of risk in the construction of super-high-rise buildings. By integrating fault tree analysis, accident chain analysis, and complex network theory with a large-scale database of 1,240 construction accidents, this study advances risk coupling research with both methodological innovation and robust empirical grounding, offering a systematic framework for safety management in super-high-rise construction.
Zhou et al. (Wed,) studied this question.