Abstract: In view of the hard problem that the weights of decision makers and criteria are usually vague and imprecise in group decision making process,we propose a linear method of multi-attribute group decision making with complete ignorance of weight information,with the emphasis on how to objectively determine the weights of decision makers and the weights of criteria only by decision matrices.We firstly introduce information entropy and evidence distance to determine the weights of decision makers,and then combine the weight vector of decision makers with the entropy weight matrix of criteria by the principle of fuzzy transformation to obtain the united weights of criteria in group decision making.Finally a linear weighted method is utilized to aggregate individual opinions of decision makers for rating the importance of alternatives.Two numerical examples for supplier selection and some relevant discussion are given to examine the feasibility and validity of the presented approach,which is characterized by clear concept,moderate computational complexity,strong objective,and easy machine implementation.In the end we extend it to hybrid multiple attribute group decision making with attribute values in the various forms of precise numbers,linguistic terms,i ntervals,and intuitionistic fuzzy numbers

Key words: multiple attribute, group decision making, weights, information entropy, evidential distance