本研究之目的在於討論通勤者在其決策過程中所出現之模糊特性,利用模糊理論試圖解釋決策過程中的不確定性;本研究將通勤行為之模糊特性界定在兩個主題:1.決策屬性認知之模糊性,2.偏好感受之模糊性,期望透過模糊理論之應用,對於通勤決策所面臨之模糊決策環境加以瞭解。本研究以淡江大學教職員生日常通勤者為研究案例,進行通勤資料之取得,其中可分為兩部份:第一部份是以敘述性偏好方式配合模糊集合理論設計問卷,取得受訪者日常通勤之靜態模糊決策行為資料,用以分析整體一般性之模糊行為;第二部份是以控制實驗方式,以真實的通勤者在模擬的通勤環境下,進行連續數日的決策實驗,作為分析實際通勤認知感受模糊特性之資料基礎。實證結果顯示:1.通勤者在決策過程中對於屬性認知的確會有模糊性產生,並以梯形函數描述較為適當;2.對於不同決策屬性亦有不同的模糊描述方式;3.進一步以蒙地卡羅模擬為基礎,結合β函數以更一般化方式測試構建屬性認知隸屬函數,改進了梯型函數間斷連續之函數型態,並更能捕捉到模糊認知中較為極端的部份;4並以模糊理論為基礎,建構出發時間變換之模糊門檻,此結果可供有關出發時間方案定義之理論基礎。
The main purpose of this study is to discuss the fuzziness in commuting decision process utilizing basic fuzzy set theory. To capture drivers' fuzzy characteristics during their commuting processes, a controlled experiment approach is implemented in this study where a group of selected auto-driving commuters interact with a simulated commuting context with repeated observations. The experiment comprises of two stages. In the first stage, every participant is presented in a preliminary survey regarding his/her perception of choices and route (link) attributes. This includes the best description of how the decision on each day was made routinely and the perception of attribute variables. Further study in the second stage is repeated simulated commuting outcome with regard to participants' daily individual departure time and route choice decisions. The results show that the fuzziness in commuting process is indeed likely and constructing fuzzy membership by trapezoid-shaped function is a proper way to describe the fuzziness.