Multiple Group Decision Support Research With Different Targets

With the rapid progress of science and technology in contemporary society and the continuous development of the knowledge, many problems become more and more huge. Decision-making problems depending on individual decision maker become less and less, but the most are required by a number of decision makers to make decision, which is regarded as the group decision-making problems. Faced with this social context, decision-making problems in the process of production and operation gradually tend to group decision-making. For example, due to the most cutting-edge, multidisciplinary, random factors and fuzziness increasing, lack of knowledge and experience, information and ability, the decision-making of medium-sized enterprises and projects, should be made by multiple experts in order to ensure the accuracy and efficiency of the decision-making and avoid errors. The decision needs professional experts and scholars with various disciplines and fields, to improve the accuracy of decision-making, reduce the possibility of serious consequence. However, because policymakers have different social status, power, prestige, knowledge and experience, preferences, expectations, etc, so the decision results may not be made exactly the same. The factors that influence the result of group decision making and the process of multi-group decision making are introduced.Firstly, because of the different backgrounds of the decision makers, the evaluation methods of decision makers are different. The different types of preference information should be transformed into the same form. In this paper, the different forms of preference information are translated into fuzzy set over the basic linguistic set and converted into two-tuple form, which transforms the different preference information into the same form.Secondly, the weights of the attributes in every group are determined. In multi-group decision making, the determination of attribute weight is one of the important research contents. The method of determining attribute weights is divided into two categories: the subjective weighting method and the objective weighting method. In order to overcome the subjectivity, this paper uses the entropy weight method to determine the weights of the attributes.In addition, the weights of experts are determined in every groups. The determination of the weights of experts is another important research point in group decision making. In the former research of group decision making, it is assumed that the expert weight is equal, or the subjective method and the objective method are adopted to obtain it. In this paper, in order to overcome the subjectivity, the two objective methods are adopted to determine the expert weights: the method based on minimizing the uncertainty and deviation degree and that based on maximizing the deviation distance and minimizing the weight distance.Finally, the opinions of multiple groups are aggregated. Two steps are involved for aggregating the opinions in each group and across every groups. Step one, the opinions of experts are collected in each group by means of simple weighted sum method. Step two, about aggregating the alternative rankings from every groups, in this paper, a distance function is defined in order to figure out an alternative ranking whose distance with other rankings form every groups is minimized by means of genetic algorithm.