基于模糊交叉效率的数据库选择研究

Research on the Selection of Database Systems Based on the Fuzzy Cross-efficiencies

交叉效率模型下决策单元自评值最优时的权重组合存在非唯一性，造成他评值不同，影响到综合效率值和排序结果。区间值构成交叉效率值，又过于模糊，增加了决策的不确定性。因此，针对权重组合的多重最优解，提出基于犹豫模糊集的交叉效率群决策方法。在原有决策单元下构建正负理想决策单元，选取从“仁慈“到“进取”的5个模型代表由悲观到乐观过渡的决策者，用犹豫模糊元表示群决策下的交叉效率值，进而构建犹豫模糊交叉效率矩阵；引入多目标优化方法确定属性的权重：改进贴近度公式，计算决策单元与正负理想决策单元的相对贴近度并进行排序。最后通过数据库系统的选择实例说明提出的决策方法是合理和有效的。

With respect to the existence of the non-uniqueness, the choice among weight sets causes different cross-efficiency scores and rankings, and the interval cross-efficiencies increase the complexity of decision making information. Therefore, a group decision making method is proposed on the basis of the hesitant fuzzy sets. The new method exploits the information of different optimal solutions. First of all, the positive and negative ideal decision-making units are introduced, and the ＂Aggressive＂ to ＂Benevolent＂ five models are chosen as decision makers from pessimism to optimism. Then, a new cross-efficiency matrix is con- structed, and the multi-objective optimization method is developed to determine the attribute weights. In ad- dition, the relative closeness degree is obtained to rank the alternatives. Finally, a numerical example about the selection of database systems is provided to illustrate the rationality and effectiveness of the proposed model.