Aggregating Experts’ Fuzzy Opinions for Hierarchical Group Decision-Making Problems

• Main Authors: Yu-TingLiu, 劉宇婷
• Other Authors: Liang-Hsuan Chen
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/28372033188847940723

碩士 === 國立成功大學 === 工業與資訊管理學系碩博士班 === 101 === In the real world, decision makers are often faced with complicated decision problems where many attributes and alternatives are considered. An alternative with respect to an attribute can be assessed by crisp data, but decision making usually includes subjective opinions. Experts can use linguistic terms to express their real opinions, and these tend to be fuzzy rather than precise. As a result, various fuzzy multiple attribute decision making (FMADM) methods, which are suitable for group decision making (GDM) problems in a fuzzy environment, have been developed. In this field, we consider a heterogeneous group of experts in a hierarchical structure, and develop a rational fuzzy opinion aggregation model. In the proposed model, an aggregation technique for a hierarchical group of experts is employed which deals with experts’ fuzzy opinions about the subjective attributes of a decision problem. The aggregation model includes three major states, the initial state, first state and second state. The purpose of the initial state is to form a committee of experts, then identify attributes and alternatives. In the first state, each expert gives performance ratings about alternatives with regard to each subjective attribute. After rating, the fuzzy data should be converted into a standardized form so that it can be calculated later on. In the second state, an aggregation method for group of experts is employed. Two factors are considered in this method. One is the weight of each expert, and it is different for each attribute. Another is the similarity of the opinions between experts. Finally, we combine these two factors linearly and find the aggregation coefficient, which helps to obtain the aggregation result of the fuzzy opinions. To verify the rationality and feasibility of the proposed method, two cases are used to demonstrate every step in each state. The results show that the proposed model is flexible and can be applied widely to different group decision making problems.