中文摘要：

对每个属性下不同决策者权重不同的决策问题，提出一种始终基于群体理想解的新方法．首先，对每个备选方案定义群体理想解和负理想解，求出对每个决策者而言该方案与其群体理想解的相对贴近度；然后，基于群体贴近度矩阵定义整个备选方案集的群体理想解和负理想解，进而计算决策群体对各备选方案的最终排序结果；最后，以供应商选择算例说明其有效性．该方法将逼近理想解法扩展到群决策，以群体理想解取代群体效用，避免了群体效用归并难题．

英文摘要：

A ranking method always based on group＇s ideal solution is proposed for a decision-making problem considering weights of decision makers with respect to each attribute. For each alternative, at first, the group＇s positive ideal solution （GPIS） and the group＇s negative ideal solution （GNIS） were defined and thus the relative closeness of it to the corresponding GPIS for every decision maker was computed. The GPIS and the GNIS for the alternative set were derived again from all the relative closenesses above and the final rankings of the alternatives for the group were determined according to their relative closenesses to the latter GNIS. It is illustrated by a supplier selection example that the proposed approach is effective. This proposed method extends the technique for order preference by similarity to ideal solution （TOPSIS） to group decision-making, substitutes group＇s ideal solution for group utility and avoids the conundrum of aggregating individual utility to group utility.