群决策中基于语言评价信息的TOPSIS方法

针对基于语言评价信息的群决策问题,提出了一种新的TOPSIS方法。依据传统的TOPSIS方法的基本思路,首先将语言决策矩阵转换为规范决策矩阵;其次,通过变量转换关系对有序语言短语集中的有序语言短语进行"量化",构造与原语言决策矩阵"等价的导出"决策矩阵,并利用算术加权平均算法对所得到的导出决策矩阵进行群的集结;然后对所得到的决策矩阵构造理想点和负理想点,计算每个方案的相对接近度,根据得到的相对接近度对方案进行排序从而确定最优方案。最后通过一个算例说明了该方法的有效性和实用性。

TOPSIS method for group decision-making based on linguistic assessment information

With regard to group decision-making problems based on linguistic assessment information, in this paper a new TOPSIS method is proposed. According as the traditional TOPSIS method, the linguistic decision matrices are transformed into normative decision matrices based on the related concepts of ordered linguistic terms set primarily, then the ordered linguistic terms are "quantized",subsequently the "equivalent" matrices to the original linguistic decision matrices are got by variable transformation,the normal matrices are aggregated by the weighed average method, and the optimal alternative is determined by calculating the distance between every alternative and ideal solution or negative ideal solution. Finally a numerical example is used to illustrate the use of the proposed method.