模化不同旅次長度之運具選擇行為

Abstract

運具選擇行為會隨著不同的旅次長度而改變，政府在進行政策管理前要先了解旅運者在不同旅次長度使用運具的偏好，才能進行政策的制定，因此本研究目的在分析不同旅次長度下大臺北都會區旅運者之運具選擇行為，並納入未來台北市政府可能實施之里程計費方案進行市佔率及政策分析。 本研究回顧過去文獻並確認模式中所需之解釋變數以及使用之研究方法，再進行問卷設計以及問卷調查計畫。使用多項羅吉特模式分析時為了突顯旅次長度對運具選擇之影響，嘗試並比較不同形式之旅次長度變數：(1)TL/TC、(2)TC〖(TL⁄¯TL)〗^λ、(3)TL分別為類別變數以及連續變數，結果當旅次長度為類別變數時模式結果較佳。將旅次長度為連續變數之結果進行巢式羅吉特模式推估並分析，發現運具方案之私人運具(機車、汽車)間具有相關性，其中一種運具選項為捷運(捷運、捷運與公車間轉乘、捷運與其他運具間轉乘)也具有相關性。 結果顯示女性、年齡愈大、教育程度愈高的旅運者相較於其他運具較不會選擇機車；所得愈高、有汽車駕照、家戶持有車輛愈多、家戶未成年人數愈多之旅運者相較於其他運具比較會選擇汽車。市佔率分析結果顯示機車隨著旅次長度增加其市佔率減少，不論在何旅次長度時其市佔率在所有運具方案中皆最大。公車、捷運及汽車之市佔率皆隨著旅次長度增加而上升，三種運具中當旅次長度在8公里以內時，公車之市佔率最大；當旅次長度大於8公里則由捷運之市佔率最大。最後納入可能之4種里程計費費率方案分析不同旅次長度下旅運者運具選擇行為的改變，其中費率方案1之起程費率最大、可搭的起程里程最長、續程費率最大，結果顯示實施費率方案1可提升較多公車使用量且降低私人運具使用量。

Mode choice behaviors may be significantly different for various trip lengths, and it is important for government to understand travellers’ preference of mode before they try to fomulate policies. Based on this, this study aims to examine the effect of fare rate and trip length on mode choice behaviors in Taipei metropolitan. To do so, this study first defines the model specification, including the adopted models and potential explanatory variables based on a systematic literature review. Then, this study designs questionnaire and survey plan to conduct a large-scale survey. To highlight the direct and compound effects of trip length on mode choice behaviors, various forms of trip length variables such as (1)TL/TC, (2), (3)dummy or continuous variable of TL, are attempted and compared. Multinomial logit model (MNL) and nested logit models (NL) with various nested structures are then estimated. Accroding to the estimated NL model, there are strong correlation among mode alternatives of car and motorcycle, as well as MRT, MRT+Bus and MRT+others. The estimation results show female, older, high educated travelers tent to less choose motorcycles; while, high income with driving license, more number of cars owned and number of minors in the household, prefer using cars. Based on model estimation results, this study analyzes mode shares that the higest mode share of motorcycle by most trip lengths, and it decrease by increasing trip length. On the contrary, mode shares of bus, car and MRT increase by increasing trip length, when trip length under 8 kilometers, the mode share of bus is the highest of the three; if over 8 kilometers, the mode share of MRT is the highest of the three. Finally, this study evaluates the changes in mode shares with four different distance-based fare schemes, and fare scheme 1 is the best for bus distance-based fare which can improve more usage of bus and decrease usage of private transportation.