两阶段灰色综合测度决策模型与三角白化权函数的改进

Two stages decision model with grey synthetic measure and a betterment of triangular whitenization weight function

首先将中心点三角白化权函数中对应于灰类1和灰类s的三角白化权函数分别取为下限测度白化权函数和上限测度白化权函数.这一改进避免了将各聚类指标的取值范围向左、右延拓的困扰.针对灰色聚类系数向量δi的各分量均衡取值或灰色聚类系数向量δi有若干个位于前列的主分量取值相近,难以判定决策对象归属的问题,构建一种新的两阶段灰色综合测度决策模型,以解决灰色聚类系数向量δi的各分量取值趋于均衡或δi有若干个位于前列的主分量取值相近情形下的综合决策问题,并通过应用实例验证了模型的有效性.

Firstly, the triangular whitenization weight function which corresponds to class 1 is changed to a whitenization weight function of lower measure, and the triangular whitenization weight function which corresponds to class s is changed to a whitenization weight function of upper measure. The puzzle of extending the bound of value of each clustering index is avoided due to this betterment. A novel two stages decision model with grey synthetic measure is proposed for the problem that it is difficult to determine the ascription of decision object at the case of each ponderance with equilibrium value of the grey clustering coefficient vector ~ or the case of some front main ponderances with close value of the grey clustering coefficient vector tJi. Finally, two practical examples are presented to verify the effectiveness of the proposed model.