號誌化道路路段旅行時間推估之研究

機構典藏 > College of Business and Management > Graduate Institute & Department of Transportation Management > Proceeding > Item 987654321/22312

Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/22312

Title:	號誌化道路路段旅行時間推估之研究
Other Titles:	A Study on the Estimation of Arterial Link Travel Time
Authors:	胡守任;陳寶如
Contributors:	淡江大學運輸管理學系
Keywords:	旅行時間;迴歸模式;類神經網路模式;時間數列模式;車輛偵測器;Travel time;Regression model;Artificial neural network model;Time series model;Vehicle detector
Date:	2005-11
Issue Date:	2009-11-30 14:01:59 (UTC+8)
Publisher:	臺北市 : 中華民國運輸學會
Abstract:	間推估模式，期望能提供完整的用路人資訊，以利用路人進行相關旅行決策之參考。本研究利用迴歸模式、類神經網路與時間數列等模式，以號誌化道路路段為研究對象，並利用微波雷達式車輛偵測器蒐集實際車流資料，據以推估道路路段的旅行時問，並針對不同模式的結果進行績效比較和分析。研究結果顯示，線性迴歸模式並不適用於號誌化道路路段旅行時間推估，其校估的R-squared值均不高，顯示流量、佔有率和速率三者問可能存在非線性的關係。至於在類神經網絡模式與時間數列模式的績效表現方面，透過相關統計檢定，兩者表現皆良好，根據學理推論與初步實證分析結果顯示，兩者均過用於號誌、化道路路段旅行時間之推估。 regression model, artifìcial neural network model and time series model to provide various analytical estimation results of arterial link travel times. The models were constructed and evaluated through field data collection using a microwave vehicle detector system. The empirical results indicated that linear regression model does not perform well enough in view of the low R-squared values, which implies nonlinear relationships among traffic variables of traffic flow, accupancy and speed might exist. Meanwhile, both artificail neural network and time series models are capable of the estimation of the present research topic. It is then concluded that, based on the theoretical background and preliminary field investigation, both models are suitable for the estimation of arterial link travel time.
Relation:	中華民國運輸學會第二十屆學術論文研討會論文集 (第三冊)=Proceeding of the 20th annual conference for the Chinese institute of transportation v.3，頁845-870
Appears in Collections:	[Graduate Institute & Department of Transportation Management] Proceeding

Files in This Item:

File	Size	Format
index.html	0Kb	HTML	234	View/Open

Loading...