考虑决策单元竞争合作动态变化的DEA博弈交叉效率方法

发布时间：2018-06-27 16:31

本文选题：DEA + 博弈　；参考：《运筹与管理》2017年05期

【摘要】：利用DEA方法进行相对效率评估时,决策单元通常需要考虑多重目标,且随着目标的变化,决策单元间竞争合作状态也会发生动态变化。传统竞合模型虽然考虑了决策单元间竞争与合作同时存在的现象,但忽视了竞争合作关系动态变化的过程。本文以竞争合作对策为切入点,将多目标规划中的优先因子引入传统DEA博弈交叉效率模型中,提出了带有优先等级的多目标DEA博弈交叉效率模型,即动态竞合博弈交叉效率模型。该模型充分体现了不同目标下决策单元间竞争合作关系的动态变化,其焦点由传统竞合模型对多重最优权重现象的改善,转向对最优效率得分的直接寻找。利用DEA动态竞合博弈交叉效率模型,本文对环境污染约束下2014年长三角地区制造业投入产出绩效进行了客观的评估。分析结果表明:DEA动态竞合博弈交叉效率模型收敛速度优于传统DEA博弈交叉效率模型,其交叉效率得分收敛于唯一的纳什均衡点;不同目标重要性的差异程度,对最终排名结果不产生明显影响,不需要确切指出。
[Abstract]:When using DEA to evaluate relative efficiency, decision making units usually need to consider multiple objectives, and with the change of objectives, the competitive cooperation among decision making units will also change dynamically. Although the traditional competition model considers the phenomenon of competition and cooperation among decision making units, it ignores the process of dynamic change of competition and cooperation relationship. In this paper, the priority factors in multi-objective programming are introduced into the traditional DEA game cross-efficiency model, and a multi-objective DEA cross-efficiency model with priority level is proposed. That is dynamic competing game cross efficiency model. The model fully reflects the dynamic change of the competition and cooperation relationship between decision making units under different objectives. The focus of the model is from the improvement of the traditional competing model to the improvement of the phenomenon of multiple optimal weights, and to the direct search for the optimal efficiency score. By using the DEA dynamic competition game cross efficiency model, this paper evaluates the input-output performance of the manufacturing industry in the Yangtze River Delta region in 2014 under the constraint of environmental pollution. The results show that the convergence speed of the cross efficiency model of the dynamic competing game in the proportion DEA is better than that of the traditional cross efficiency model of DEA game, and the score of cross efficiency converges to the unique Nash equilibrium point. Does not have the obvious influence to the final rank result, does not need to point out exactly.
【作者单位】：南京信息工程大学经济管理学院中国制造业发展研究院;
【基金】：国家自然科学基金资助项目(71571104) 江苏省自然科学基金资助项目(BK20141481)
【分类号】：O221

【相似文献】