Proposal of aggregation technique for multiple criteria decision making problems

Multi criteria decision making techniques are utilized in solving of problems encountered in different areas. Decision making problems encountered are grouped into three groups: selection, ranking and classification. There are many multi criteria decision making techniques that can be used to solve problems in these three different setup. However, the results obtained with these methods may differ from each other. At this point, it is ambiguous through which method the best result can be achieved. A recent increasingly favored approach is, therefore, to attain solutions via different multi criteria decision making models that are compatible with problems and reaching a final solution by aggregating these solutions in the last phase. Techniques that are used in aggregating solutions obtained through different multi criteria decision making models are called as aggregation techniques. With the purpose of aggregating ranks, this study offers the Reference Alternative Based Aggregation Technique (RAT). The main goal of RAT is to aggregate results of different multi criteria decision making models which are used for a problem. It differs from other aggregation techniques in that it is easy to use, does not require an additional technique or scoring, and largely eliminates the problem of complete ranking. Analysis has shown that RAT gave 99.93% compatible results with other aggregation techniques. It was, however, observed that complete ranking could not be obtained in an average of 1.2% of the results with RAT. On the other hand, it is observed that, compared with RAT, this rate was at least 82% for other techniques. RAT was applied on 830,000 samples and the results were compared with Borda, Copeland, Dodgson and Kemeny aggregation techniques. Furthermore, previous 21 studies that used aggregation techniques were reanalyzed with RAT and the results were compared. In conclusion, this study suggests that RAT will make important contributions to the field of multi criteria decision making and can be successfully applied in areas such as social preferences and group decision making.