Abstract: Based on AHP, the qualitative and quantitative risk analysis method based on AHP-the grey system theory-artificial neural network-finite element analysis-Monte Carlo method was proposed. And the proposed method was applied to a long-span rail-cum-road cable-stayed bridge with single cable plane and steel truss girder. The research results indicate that, when the number of structure displacement state simulation is more than two millions times, the steel truss cantilever end displacement failure probability reaches 3%, and the failure probability is in control. The proposed risk analysis method using probability theory more accurately determines the impact of bridge risk factors.

Key words: brideg engineering;construction risk analysis, analytic hierarchy process, cable-stayed bridge, artificial neural network, Monte Carlo method, quantitative analysis;single cable piane

摘要： 在层次分析法的基础上提出了基于层次分析法-灰色系统理论-人工神经网络-有限元分析-蒙特卡罗理论的定性、定量风险分析方法，并将其应用于某大跨单索面公轨两用钢桁梁斜拉桥。研究结果表明：当结构位移状态仿真次数超过200万次时，钢桁梁最大悬臂端位移失效概率仅为3%，失效概率可控，该风险分析方法采用概率理论更加准确的确定了桥梁风险因素对桥梁结构的影响。

关键词: 桥梁工程, 施工风险分析, 层次分析法, 斜拉桥, 人工神经网络, 蒙特卡罗理论, 定量分析, 单索面