在運輸規劃領域中,規劃者常需要有起迄旅次資料作為輔助,藉此瞭解旅次分佈的情形後,進而針對運具分配及交通指派等方面進行更適切的規劃。但以往取得起迄旅次資料的方式,通常需要大量的時間、金錢以及人力資源,且常因為天候或人為、機器疏失的影響,造成所取得之起迄旅次資料準確度不高,因此,利用模式推估起迄旅次資料,是運輸相關研究中持續進行的課題。針對運輸廊道而言,由於各起迄點之間的指派關係不易取得,故研究常以匝道進出口流量作為推估時所依據的資料,而如何利用進出口流量建構各起迄點之間的關係,是模式中最重要的課題。 本研究以Tsygalnitzky(1977)所提出之管流類推法為基礎,延續郭衣宬(民94)與林志豪(民95)之研究,持續針對管流類推法應用在封閉型運輸廊道上之起迄旅次推估進行探討;另外,研究中也針對以往該法在應用中所做之假設進行討論,包括全天採用單一相對吸引力值、匝道間距相同等等。因此,本研究重新建構之依時性管流類推法修正模式,其相對吸引力值將隨時間變動,亦考慮模式中進出口相關資訊間之時間因素。透過模擬案例的測試,可發現模式在處理具時間向度之不同流量變化,具有不錯的推估能力。結果顯示加入依時性之修正,對於管流類推法應用於封閉型運輸廊道之起迄旅次推估,能有效提升其推估結果。
In the field of transportation planning, planners often need some origin- destination data to know about the trip distribution in the region, and then do more properly planning in modal split and traffic assignment. However, the traditional ways which obtain the O-D data usually cost a large amount of time, money, and human resources. In addition, since it could be affected by weather, artificially negligence or mechanical disorder, the O-D data we obtained might have low accuracy. Therefore, the research on the model in O-D estimation has been a continuous study topic in transportation. Due to the assignment between each O-D pair in corridor is hard to obtain, so the data set used by estimation in research are often the in-flow and out-flow counts of access ramps, and how to use these flow data to build the relationship between each O-D pair is very important in the model. This research is based on fluid analogy method proposed by Tsygalnitzky, following the research of Kuo(2005) and Lin(2006), continued to discuss the development and application of fluid analogy method on O-D estimation of access-controlled corridor. The hypothesis used in the past, including the unitary factor of relative attractiveness for all day long and the fixed distance among ramps were re-examined in this research. For the modified model in this study, the fixed factor of relative attractiveness in the time-dependent fluid analogy method varies with time, and so do the distances among ramps. According to the simulation tests, the model was able to deliver sound results in the situation of flow varying along time dimension.