Abstract: In order to prevent and control the conflict of technical interface of railway construction in difficult and dangerous mountainous areas, the WSR-K-Means++ model was proposed to identify the influencing factors of technical interface conflicts of difficult and dangerous mountain railways construction and classify them. Based on the methodology of the WSR system, the influencing factors of technical interface conflicts in difficult and dangerous mountainous railways construction were identified; the four-dimensional attribute membership degree of each influencing factor was defined and calculated from the perspective of four-dimensional attributes such as technology, responsibility, time and space. And Matlab was used to simulate the K-Means++ algorithm to cluster the membership. Taking a railway bid section in a difficult and dangerous mountainous area in southwest China as an example, the rationality and practicability of the WSR-K-Means++ model was verified. The results show that 80 factors affecting construction technology interface conflicts are identified, and the influencing factors should be divided into 4 levels. Through hierarchical management of influencing factors, it has certain practical value and reference significance for the timely prevention and control of technical interface conflicts in difficult mountain railways construction and the improvement of management efficiency.

Key words: railway engineering; difficult and dangerous mountainous areas; technical interface of construction; influencing factors; WSR; clustering analysis; conflict

摘要： 为预防和控制艰险山区铁路施工技术接口冲突，提出了WSR-K-Means++模型识别艰险山区铁路施工技术接口冲突影响因素并划分其等级。以WSR系统方法论为基础，识别艰险山区铁路施工技术接口冲突影响因素；从技术-责任-时间-空间四维属性角度定义并计算各影响因素四维属性隶属程度，并运用MATLAB模拟K-Means++算法对隶属度进行聚类分析；以西南艰险山区某铁路标段为例，验证WSR-K-Means++模型的合理性和实用性。结果表明：识别了80个施工技术接口冲突影响因素，影响因素应划分为4个等级；通过对影响因素分层管理，对艰险山区铁路施工技术接口冲突及时防控及管理效率的提高具有一定的实用价值和参考意义。

关键词: 铁道工程；艰险山区；施工技术接口；影响因素；WSR；聚类分析；冲突