IPOWA算子及其在正态随机多准则决策中的应用

发布时间：2018-04-24 09:01

  本文选题：多准则决策 + 正态随机变量　； 参考：《计算机科学与探索》2017年01期

【摘要】：现实中,很多现象近似服从正态分布。为了解决准则值为正态随机变量且决策者风险态度不确定的多准则决策问题,提出了一种基于区间可能有序加权平均(interval possible ordered weighted averaging,IPOWA)算子与可能性系数的方法。通过正态随机变量的线性组合方法,算出各方案的线性加权正态随机变量。根据正态随机变量的3σ原则,将线性加权正态随机变量转化为随机区间数。对随机区间数进行两两比较,得到可能度矩阵,利用区间可能有序加权平均算子,集结可能度矩阵中每行元素的值。采用风险态度的可能性系数,计算各方案的综合评价值,进而确定各方案的最终排序。最后,通过两个算例的对比分析,表明所提方法具有排序结果稳定,支持信息充分等特点。
[Abstract]:In reality, many phenomena approximate normal distribution. In order to solve the multi-criteria decision making problem in which the criterion value is a normal random variable and the decision maker's risk attitude is uncertain, a method based on interval possible ordered weighted averaging operator and possibility coefficient is proposed. By using the linear combination method of normal random variables, the linear weighted normal random variables of each scheme are calculated. According to the 3 蟽 principle of normal random variables, the linear weighted normal random variables are transformed into random interval numbers. Through the pairwise comparison of random interval numbers, the possibility degree matrix is obtained, and the values of each row element in the possibility degree matrix are aggregated by using the interval possible ordered weighted average operator. The probability coefficient of risk attitude is used to calculate the comprehensive evaluation value of each scheme, and then the final ranking of each scheme is determined. Finally, the comparison and analysis of two examples show that the proposed method has the characteristics of stable ranking results and sufficient support information.
【作者单位】： 湖南大学工商管理学院;湖南商学院湖南省移动电子商务协同创新中心;湖南科技大学管理学院;
【基金】：国家社会科学基金 教育部人文社会科学研究青年基金 中国博士后科学基金 湖南省哲学社会科学基金 湖南省社会科学成果评审委员会重大课题 湖南省教育厅科学研究优秀青年项目~~
【分类号】：C934
，

本文编号：1795997