基于效率的组织多属性决策及实证研究:DEA-TOPSIS组合方法

发布时间：2018-12-12 23:17

【摘要】：在日益激烈的竞争中,决策和效率已成为组织获得竞争力的关键因素,但在管理实践中往往会出现以牺牲效率来达到决策优化的目的。为了在保证效率甚至提高效率的基础上优化组织的决策,本文首次从基于效率进行多属性决策的角度,将DEA方法和TOPSIS方法进行组合。DEA方法不仅可以对具有多种投入多种产出指标的组织的相对效率进行测算,还可求得决策单元(Decision Making Unit,DMU)各指标的松弛改进量,这使得DEA方法与TOPSIS方法进行组合在理论上是可行的。以首都医科大学为例,假设组织为了提高整体的效率竞争力,希望在2013年效率的基础上使有效DMU的个数增加,并在实现过程中使各指标松弛改进总量尽可能小(即使改变尽可能"容易")。将首都医科大学附属的10所三甲综合医院作为DMUs,并运用DEA-TOPSIS组合方法,从提高技术效率的角度进行了研究。研究结果表明,DEA-TOPSIS组合方法不仅可以有效地对基于效率的决策备选方案进行排序,还可以通过选择不同的模型和指标处理方法以尽可能地反映实际情况,具有很强的实践价值。
[Abstract]:In the increasingly fierce competition, decision-making and efficiency have become the key factors for the organization to gain competitiveness. However, in management practice, the goal of decision-making and optimization is often achieved at the expense of efficiency. In order to optimize the decision-making of the organization on the basis of ensuring efficiency and even improving efficiency, this paper first starts from the angle of multi-attribute decision-making based on efficiency. The DEA method and the TOPSIS method are combined. The DEA method can not only measure the relative efficiency of the organization with multiple input and multiple output indexes, but also obtain the relaxation improvement of each index of the decision making unit (Decision Making Unit,DMU. This makes the combination of DEA method and TOPSIS method feasible theoretically. Take Capital Medical University as an example, suppose that organizations want to increase the number of effective DMU on the basis of efficiency in 2013 in order to improve their overall efficiency competitiveness. In the process of implementation, the total amount of relaxation improvement is as small as possible (even if the change is as easy as possible). In this paper, 10 hospitals affiliated to Capital Medical University were used as DMUs, and DEA-TOPSIS combination method was used to study the improvement of technical efficiency. The results show that the DEA-TOPSIS combination method can not only effectively rank the decision options based on efficiency, but also select different models and index processing methods to reflect the actual situation as much as possible. It has strong practical value.
【作者单位】：北京理工大学管理与经济学院;延安大学经济与管理学院;
【基金】：国家自然科学基金面上项目(71672011);国家自然科学基金重点项目(71432002)
【分类号】：C936;R197.32

【相似文献】