| Summary: | 博士 === 國立成功大學 === 交通管理學系碩博士班 === 97 === Multi-attribute decision making (MADM) is one of the most well known branches of decision making. Several methods have been proposed for solving related problems, but a major criticism of MADM is that different techniques may yield different results for the same problem. Group decision making is an active area of research within MADM, which attempts to aggregate individual judgments into a group judgment. However, different group preference aggregation methods will often lead to different results.
Under multiattribute group decision making, various individual preference aggregation methods and MADM approaches often lead to different outcomes for selecting or ranking decision alternatives involving multiple attributes. This suggests that the choice of a specific method will significantly influence the ranking outcome. To help managers make better decisions, a mechanism is thus required for selecting an appropriate outcome for a given MADM problem.
An empirical study, entitled “Funding Source Allocation for Public Transportation Subsidy”, was conducted to explain how the proposed approach can be used to help select the best ranking outcome under multiattribute group decision making. Six possible funding sources embodied by four attributes were proposed to subsidize the shortage caused by the welfare policy. AHP was used in data analysis, while two main approaches: aggregating individual judgment (AIJ) and aggregating individual priority (AIP) were used for aggregating information, and SAW, WP and TOPSIS were used for scoring phase. Nine methods were finally used to allocate funding sources. The results show that the method “AIJ and geometric mean for aggregation and WP for the scoring phase” is the best, since its results have the highest degree of consistency degree with experts.
Another empirical study, entitled “Green Bus Technology Selection”, was conducted to explain how the proposed approach can be used to select the best ranking outcome under fuzzy multiattribute group decision making. Six possible green bus technologies embodied by six attributes were proposed to select the most appropriate one for Taiwan. Arithmetic and geometric means were used to integrate the fuzzy judgment values of evaluators. SAW, WP and TOPSIS were used for the scoring phase and three defuzzification methods were used to convert the fuzzy data into crisp scores. Eighteen methods were finally formed for solving the problem.The results show that the method “arithmetic mean for aggregating the fuzzy judgment, SAW for MADM phase and center-of-area method for defuzzification” is the best, since it has the highest degree of consistency with experts. Finally, different MADM problems and data sets may lead to the selection of a different method.
|
|---|
