Abstract: As the complexity and uncertainty of the tunnel engineering, there is a considerable risk in the entire construction process. After investigating many history cases, this paper sums up these cases from four aspects, including question, scene, solution and result. To improve the efficiency and accuracy of case retrieving in case-based reasoning, the rough-set theory is introduced to reduce case attributes and get attribute weights. Compared with traditional technique, this technique reduces the subjective effects and improves risk assessment of the accuracy and efficiency, and guarantees the safety of the tunnel construction.

Key words: rough set, tunnelling, risk, case-based reasoning

摘要： 隧道工程具有复杂性和不确定性，其施工过程存在很大风险。该文调研隧道工程历史案例，从问题、情景描述、求解方法和结果4个方面进行案例库的设计和归纳。根据粗糙集理论，利用差别矩阵实现对案例决策表特征值属性约简和属性权重的学习。与传统方法相比，该方法有利于减少主观影响，提高风险评估准确率和效率，保证了隧道施工的安全性。

关键词: 粗糙集, 隧道施工, 风险, 案例推理

CLC Number: 

• U455