Abstract In many decision making environments, group decision making is a necessity. in these problems, There are several criteria and alternatives and decision makers should eva‎luate the alternatives on these criteria. Decision makers may use various formats of information to eva‎luate alternatives due to different knowledge and experience they have. These kinds of multi attributes group decision making problems are called heterogeneous magdm problems. Assessment information can include real numbers, interval numbers, linguistic labels and fuzzy sets. Fuzzy set theory is an extension of ordinary set theory. Fuzzy include all kind of vagueness and uncertainty, especially the vagueness of human thinking and linguistic language. So, using the fuzzy sets can encounter with the vagueness of human being thinking by mathematical tools and handle the way of human being thinking in decision making. The fuzzy sets have different sorts and can be used in eva‎luating the alternatives due to the knowledge of decision maker. Furthermore, in real world there are feed backs between attributes and considering these relationships are important in making a more reliable decision. Besides, decision support systems can enhance the process of decision making due to their specifications and pertaining to computers. So integrating both domain of decision support systems and multi attributes group decision making can present a powerful decision making tool. The aim of this research is to design an efficient group decision making tool to provide decision makers with using different formats of information including real numbers, multiple linguistic lables, triangular fuzzy numbers, intuitionistic fuzzy sets, interval fuzzy sets and interval valued intuitionistic fuzzy sets to eva‎luate the alternatives in such a way that can handle the feedbacks among the attributes by the both methods of fuzzy DEMATE and adaptive fuzzy ANP. In adaptive fuzzy ANP, the consistency of fuzzy pair wise decision matrix can enhance by using genetic algorithm which is a huge leap in making a reliable decision. Consistency is an important factor which you can’t rely on pair wise judgments without the existence of this factor and by increasing in the number of criteria and alternatives doing the consistent pair wise judgments need more time and accuracy. So the method presented in this research provide the possibility of reaching a consistent fuzzy pair wise matrix in the least of time and because of its fuzziness can deal with the vagueness of the human being thinking. In this method, firstly, the pairwise decision matrix should be constructed by decision maker. Then, the contradictory of this matrix will be checked and in the case of the lack of this factor, the consistency of this matrix will be improved by using genetic algorithm. So, it wouldn’t be necessary to ask the decision maker to change the first decision matrix which he constructed. In this algorithm two aims will be considered. The first one is improving the consistency of the decision matrix and the second one is having the minimum distance from the decision maker opinion. Besides, due to huge amount of calculations which this methodology has, we design a group decision software by java programming which can give the benefit of flexibility due to the classes and methods it has and it can be extended to web environment to provide the possibility of making decision from different places. Finally, this method will be compared with fuzzy TOPSIS and to show the efficiency of this system is used in a lamp manufacturing industry to select the best supplier. Key words: group decision support system, fuzzy DEMATEL, adaptive fuzzy ANP,consistency, genetic algorithm.

بهروز ارباب شيراني , مهدي علينقيان

بهروز ارباب شيراني , مهدي علينقيان