城市轨道交通运营效率评价——基于改进的博弈交叉效率方法

本文通过对有效和非有效决策单元施加限制或引入正则限制条件下提出了三种改进的博弈交叉效率评价方法,与传统博弈交叉效率评价方法相比,改进方法不仅利于在相同的权重空间中对不同决策单元进行比较,而且在收敛速度上有明显改善,在一定程度上可消除传统博弈交叉效率结果的不惟一问题.结果表明对所有决策单元施加限制可提高对决策单元的区分度,从而改善结果的不惟一性,施加全局限制可提高迭代算法的收敛性.利用改进模型对中国9个城市的轨道交通企业运营效率进行了评价,评价结果也表明改进方法在区分决策单元上有更好的表现,收敛速度占绝对优势,得到了更为可信的评价结果.

Evaluating the operational efficiency of rail transit: An application of improved game cross efficiency approach

Aiming at the possible multiple solutions and slow convergence rate in the game cross efficiency method, researchers proposed three improved approaches for game cross efficiency evaluation including restrictions on DEA-efficient units and non-DEA-efficient units, or meeting the same normalization constraint. Compared with the traditional game cross efficiency evaluation method, the improved approaches can be utilized for assessing the performances of the DMUs on a common basis, can improve the convergence rate of algorithm, and can eliminate the non uniqueness in the traditional game cross efficiency method. The result shows that the constraint of all the decision making units can be eliminated by improving the discrimination of the decision units, the application of normalization constraint can improve the convergence rate of the iterative algorithm. The proposed new approaches are illustrated with rail transit enterprise of 9 city operation performance, the results also show that the improved approaches are better in discrimination and convergence rate, and we obtained a more credible evaluation results.