由於傳統之多屬性決策理論無法對評判值之趨向自然性有效解決，於是模糊多屬性決策理論便被提出來解決自然性評判值的問題。模糊屬性決策有兩大重要部份，一為模糊綜合評判(fuzzy decision making) ，男一為模糊選序(fuzzy ranking) ，本文即是針對現有模糊選序方法作一些研究，希望能找到已知模糊選序方法之缺點及其適用時機。由本研究探討得知， Yager法十分不適丹於三角形，至於， Bass法則是三角形梯形都差不多，若不是非常需要高準確性之下， Bass法之簡單易懂，實作容易，也是一個不錯之選擇，
Chen&Hwang法雖有「反直覺」和「斥直覺」之缺點，但是在一般量化中， 卻達百分之百之正確率，因此，此法可兼顧簡單易懂、實作容易與高準確性，至於Lee&Li法雖然整體效能相當好，但計算Chen&Hwang法繁複許多，且無法適用於風險趨向者，故和Cher地Hwang法本研究認為不分軒輊。 The nature of decision making is towards fuzziness. Lacking of solving fuzziness nature of
decision making value by traditional multiple attributes decision making, fuzzy multiple attributes decision making has been suggested. Fuzzy multiple attributes decision includes two major parts, one is decision making and another is fuzzy ranking. In this paper we want to evaluat all the ranking methods and discuss their weakness and strength. According to research results, we found that Yager's method very unsuitable for triangle fuzzy number. Bass's method is not very well both for the triangle and trapezoid fuzzy number, yet it is easy to understand and implement. Chen&Hwang's method has some shortcomings, such as anti-intuition and counter-intuition.; Surprisingly, it exhibits no inaccuracy in quantity analysis, and also is easy to understand and implement. Lee&Li's method obtained fairly good result is more complicated than Chen&Hwang's method and is only suited for risk-taking.
關聯:
第六屆國際資訊管理學術研討會論文集=Proceeding of the 6th International Conference on Information Management，頁197-204
