针对已有的快慢车停站方案优化模型复杂度过高，车站分级方法不合理的情况，应用灰色变权聚类模型对车站进行初步聚类形成决策属性，以各指标聚类结果作为条件属性，对车站分级影响因素进行属性约简，应用约简属性优势分析确定各条件属性、决策属性间的灰色相对关联度，并用灰色定权聚类模型对车站进行分级． 通过设定一级车站快车必须停车，二级车站快车根据优化需要停车及三级车站快车不停车的原则，依据分级结果构建快慢车停站方案优化的 0 －1 非线性规划模型，运用遗传退火算法求解模型． 结果表明，该方法可剔除大量无效解，解空间大幅缩小，求解效率得到了较大提高，对于快速制定列车运营方案具有重要的意义．

In view of the high complexity and unreasonable station classification of the existing stopping schedule optimization models of express/local trains，by utilizing the grey variable weight clustering model to preliminarily cluster stations so as to achieve the decision attributes and by taking the clustering results of each index as the condition attributes，the attribute reduction of the factors influencing the station classification is conducted． Then，the advantage analysis of the reduced attributes is performed to determine the relative grey correlation degree between the condition attributes and the decision ones，and the grey fixed weight clustering model is adopted to classify stations． By setting the principle that express trains must stop at the stations in the first level and they may stop in the second level if it is required but do not stop in the third level，a nonlinear 0 －1 programming model of the stopping schedule optimization of express/ local trains is constructed according to the classification results，and the constructed model is solved by using the genetic-annealing algorithm． The results show that the proposed method can eliminate a large number of invalid solutions and greatly narrow the solution space，thus greatly improving the solution efficiency，which is of great significance in quickly establishing the operation scheme of trains．