语言型群决策专家信息集结的随机EMD方法
|
摘要
针对现有语言型群决策方法在集结专家信息时大多存在算法主观性大、信息利用不充分的问题,文章提出专家信息集结的随机EMD方法。基于正态分布隶属函数描述专家总体认知,运用经验模态分解,通过随机模拟估计总体期望值,基于正态分布曲线的3σ原则估计总体方差,将专家信息集结为正态分布随机变量信息,结合经典灰色随机决策方法对方案排序择优。案例对比分析显示,该方法在专家信息集结时不需要计算专家权重,充分利用决策信息,决策结果稳定可靠。
In view of the problem of large subjectivity of algorithm and insufficient use of information in gathering expert information in the existing linguistic group decision making(LGDM) methods, this paper proposes a stochastic EMD method for expert information aggregation. Based on the assumption that expert collectivity concepts can be described by normal distribution membership function, the paper uses the empirical mode decomposition(EMD) and stochastic simulation to estimate the overall expected value, employs 3σ principle of normal distribution curve to estimate the overall variance value, and aggregates the experts' information to normal distribution stochastic variable information. Finally the paper combines with the classical grey random decision method to optimize the scheme ranking. The case comparison shows that in this method there is no need to calculate the expert weights in aggregating expert information, and that making full use of decision information leads to stable and reliable decision results.
In view of the problem of large subjectivity of algorithm and insufficient use of information in gathering expert information in the existing linguistic group decision making(LGDM) methods, this paper proposes a stochastic EMD method for expert information aggregation. Based on the assumption that expert collectivity concepts can be described by normal distribution membership function, the paper uses the empirical mode decomposition(EMD) and stochastic simulation to estimate the overall expected value, employs 3σ principle of normal distribution curve to estimate the overall variance value, and aggregates the experts' information to normal distribution stochastic variable information. Finally the paper combines with the classical grey random decision method to optimize the scheme ranking. The case comparison shows that in this method there is no need to calculate the expert weights in aggregating expert information, and that making full use of decision information leads to stable and reliable decision results.
引文
[1]周任军,陈瑞先,陈跃辉等.基于EMD提取专家语言评价信息的群决策方法[J].系统工程理论与实践,2016,36(3).
[2]Huang N E, Shen Z, Long S R. The Empirical Mode Decompositionand the Hilbert Spectrum for Nonlinear and Non-stationary Time Se-ries Analysis[C].Proceeding of the Royal Society of London, 1998.
[3]李海涛,罗党,韦保磊.不确定语言评价信息下大群体决策的MC-EMD方法[J].中国管理科学, 2017, 25(4).
[4]Li D Y, Liu C G, Gan W Y. A New Cognitive Model:Cloud Model[J].International Journal of Intelligent Systems, 2009, 24(3).
[5]戴跃强,徐泽水,李琰.语言信息评估新标度及其应用[J].中国管理科学, 2008, 16(2).
[6]Xu Z S. Induced Uncertain Linguistic OWA Operators Applied toGroup Decision Making[J]. Information Fusion, 2006, 7(2).
[7]Wang J Q, Lu P, Zhang H Y, et al. Method of Multi-criteria Group De-cision-making Based on Cloud Aggregation Operators With Linguis-tic Information[J].Information Sciences, 2014,(274).
[8]肖新平,毛树华.灰预测与决策方法[M].北京:科学出版社, 2013.
[9]董哲仁.试论生态水利工程的基本设计原则[J].水利学报, 2004, 35(10).
[2]Huang N E, Shen Z, Long S R. The Empirical Mode Decompositionand the Hilbert Spectrum for Nonlinear and Non-stationary Time Se-ries Analysis[C].Proceeding of the Royal Society of London, 1998.
[3]李海涛,罗党,韦保磊.不确定语言评价信息下大群体决策的MC-EMD方法[J].中国管理科学, 2017, 25(4).
[4]Li D Y, Liu C G, Gan W Y. A New Cognitive Model:Cloud Model[J].International Journal of Intelligent Systems, 2009, 24(3).
[5]戴跃强,徐泽水,李琰.语言信息评估新标度及其应用[J].中国管理科学, 2008, 16(2).
[6]Xu Z S. Induced Uncertain Linguistic OWA Operators Applied toGroup Decision Making[J]. Information Fusion, 2006, 7(2).
[7]Wang J Q, Lu P, Zhang H Y, et al. Method of Multi-criteria Group De-cision-making Based on Cloud Aggregation Operators With Linguis-tic Information[J].Information Sciences, 2014,(274).
[8]肖新平,毛树华.灰预测与决策方法[M].北京:科学出版社, 2013.
[9]董哲仁.试论生态水利工程的基本设计原则[J].水利学报, 2004, 35(10).