有序分位加权集结算子及理想激励点的模拟求解.PDFVIP

第33 卷第5 期 系统工 程 学 报 Vol.33 No.5 2018 年10 月 JOURNAL OF SYSTEMS ENGINEERING Oct. 2018 有序分位加权集结算子及理想激励点的模拟求解 李伟伟, 易平涛, 郭亚军 (东北大学工商管理学院, 辽宁沈阳110169) 摘要: 面向管理决策中的激励问题, 提出了一种新的集结方法, 有序分位加权集结算子. 该算子通过分布分位数的方 式衡量被评价对象的相对发展程度, 以分段激励的方式在集结过程中融入决策者的激励偏好, 从而使得被评价对象 对决策者的激励偏好有着更清晰的认识. 在对算子及相关参数进行介绍的基础上, 分析了激励点的确定原则, 并采 用随机模拟方法求解理想激励点. 以应用算例的形式验证了该求解方法的有效性, 并得出采用正向双点激励方式不 仅能够实现对被评价对象的奖惩并重, 且能够通过奖惩凸显被评价对象之间差异的结论. 关键词: 多属性决策; 激励评价; 有序分位加权集结算子; 分位权重向量; 理想激励点 中图分类号: C934 文献标识码: A 文章编号:2018 doi: 10.13383/ki.jse.2018.05.004 Ordered quantile weighted aggregation operator and the simulative solution of ideal incentive points Li Weiwei, Yi Pingtao, Guo Yajun (School of Business Administration, Northeastern University, Shenyang 110169, China) Abstract: This paper proposes a new aggregation method for the incentive problems in management: denominated-ordered quantile-weighted aggregation operator. The quantile variable is used to measure the relative development degree of alternatives, and partitioned stimulation is used to fuse the incentive preference of decision makers into the aggregation process. Therefore, it can make the alternatives understand more of the decision maker’s incentive preference. After introducing this operator and the related parameters, the paper discusses the principle of incentive point, and proposes a stochastic simulation method to find optimal incentive points. A numerical example is given to illustrate the validity of this method. It is found that under the situ- ation of double points and positive stimulation, this new operator can not only re