運輸廊道站間起迄量推估方法之研究

Estimation of Origin-Destination Matrixes on the Access-Controlled Corridors

郭衣宬(Yi-cheng Kuo)

指導教授 ： 廖祐君

中原大學/工學院/土木工程研究所/碩士(2005年)

https://doi.org/10.6840/cycu200500067

摘要

旅次起迄分佈為運輸規劃中之重要參考資訊，然旅次起迄調查之資料取得十分困難，且須投入大量人力資源，並可能發生資料錯誤或不完整。就進出控制(access-controlled)之運輸廊道而言，廊道上各點之進出流量通常較易藉由調查取得，但各點間之起迄關係則不易獲得。起迄推估模式的發展，一直是近幾十年運輸規劃發展的重要課題。針對複雜的都市路網，傾向以複雜起迄推估模式作為基礎，推估時需引入之歷史資訊相對較多，才能夠有效降低較複雜運輸路網(如都市路網以及區域間具支線公路路網)起迄推估的誤差。本研究主要以較單純不具支線的廊道系統為研究對象，而單純運輸廊道的起迄間的彼此影響及複雜度較不似路網，故模式發展朝向較簡化的起迄推估模式方向發展，並利用較易取得的廊道出入站流量作為基礎推估的資訊，進而將發展之模式運用於廊道運輸系統未被建立的階段，以僅有的流量資訊推估起迄矩陣，因為預先推估的起迄矩陣，其對於路廊之設計與流量管理具相當重要性。
本研究以Tsygalnitzky(1977)所提出之管流類推法(Fluid Analogy Method)為基礎，此法適用於管流封閉空間，對應於實際之運輸系統，捷運、公車及高速公路等具進出控制之廊道，其路線類似管流，分支少且單純，容易獨立出特定分析之範圍。研究中，分為三部分予以探討，第一部份就最單純之封閉型廊道，進行基本推估方法之建立及相對吸引之探討，且廊道中各出入站之進出量皆完整提供。第二部分，則放寬部分條件，允許廊道中某站之進或出資料不全，進而考慮當廊道中某站之進出資料皆缺乏，如何予以推估起迄量分別加以探討。第三部分將實際高速公路起迄量作為管流法推估基礎，並加入額外起迄資訊降低管流法推估起迄錯誤率，透過三個部分探討管流法模式的特性，並藉此得到近似真實之廊道站間起迄推估，減少實際問卷及高成本之調查工作。

關鍵字

相對吸引參數 ； 進出控制運輸廊道 ； 管流類推法 ； 起迄矩陣

並列摘要

Information of trip interchange between areas is vital in most transportation planning practices; however the OD data collecting is difficult, costly, and frequently fragmented. Compared to such complication, it is relatively easy to gather ramp inflows and outflows of an access-controlled. The estimation of trip interchange between corridor access points is different from urban road network which has open limits with zonal trip distribution estimated by conventional gravity models in use of zonal attributes. This study aims to develop a simplified estimation process of origin-destination matrixes on access-controlled corridors without branches based on limited and easily attainable data, e.g., ramp flow counts. The estimation method is also application on other transportation system with similar forms such as bus lines or regional rails.
The estimation process in this study was based on Fluid Analogy Method proposed by Tsygalnitzky(1977) and applied on closed-ended tube system. In the first part of the study, the basic estimation method was developed and the factor of relative attractiveness derived as well as systematically examined. Next, the estimation process was modified by allowing incomplete ramp flow counts and invited rectifying mechanism in the second concluded that the modified estimation method is able to generate satisfying OD estimates without use historical, complicated, or sampled data. More conclusions and suggestion for future works were stated.

並列關鍵字

origin-destination matrix ； access-controlled corridor ； factor of relative attractiveness ； fluid analogy method

參考文獻

1. Bell, M. G. H., (1983) “The estimation of an origin-destination matrix from traffic count,” Transportation Science, Vol.17B, No.2, pp.198-217.

2. Bell, M. G. H., (1991) “The estimation of an origin-destination by constrained generalized least square,” Transportation Research B, Vol.25, pp.13-22.

3. Cascetta, E., (1984) “Estimation of trip matrices from traffic counts data and survey data: a generalized least squares estimator,” Transportation Research B, Vol.18B, No.4/5, pp.289-299.

4. Fisk, C. S. H., (1988) “The estimation of an origin-destination matrix from traffic count,” Transportation Research B, No.1, pp.69-79.

5. Hazelton, M. L., (2000) “Estimation of origin-destination matrices from link flows on uncongested network,” Transportation Research B, Vol.34, pp.549-566.