研究城轨运营事故的分布特征和致因机理，对保障运营安全、制定安全管控措施具有重要意义。本文统计分析了1970~2022年国内外城轨运营事故425例，对比分析了运营事故的发生原因及时间分布特征。基于致因机理并结合主成分分析法，构建了城轨运营安全评价体系，提出了基于博弈论的组合权重评价方法，并以1990~2022年我国274例运营事故数据为例，结合专家打分情况，从宏观角度分析了我国城轨运营安全状况。研究结果表明：引发运营事故的因素有人员因素、设备因素和环境因素，其中国内外运营事故由设备原因导致的事故占比最高，分别占比65%、56%，1月、3月、7月、8月和12月为事故频发月份，与客流高峰月份相同；组合赋权法既考虑了客观统计数据中的信息量，又结合了主观专家的经验积累，使得评价结果更接近实际运营情况，也更具有说服力。

Studying the distribution characteristics and cause mechanism of urban rail transit operation accidents is of great significance for ensuring operation safety and formulating safety control measures. This paper statistically analyzes 425 urban rail transit operation accidents at home and abroad from 1970 to 2022, and compares and analyzes the causes and time distribution characteristics of operation accidents. Based on the cause mechanism and the principal component analysis method, the safety evaluation system of urban rail transit operation is constructed, and the combined weight evaluation method based on game theory is proposed. Taking the data of 274 operation accidents in China from 1990 to 2022 as an example, combined with the expert scoring situation, the safety status of urban rail transit operation in China is analyzed from a macro perspective. The results show that the factors causing operational accidents include personnel factors, equipment factors and environmental factors. Among them, 47 % of foreign operational accidents are caused by personnel factors, and 65 % of domestic operational accidents are caused by equipment factors. January, March, July, August and December are the months with frequent accidents, which are the same as the peak months of passenger flow. The combination weighting method not only considers the amount of information in objective statistical data, but also combines the experience accumulation of subjective experts, which makes the evaluation results closer to the actual operation situation and more convincing.

U298.5??????

国家自然科学基金(72261025)；甘肃省高等学校创新基金项目(2021B-103)；兰州交通大学-天津大学联合创新基金(2021056)