The Application of Fuzzy Clustering on Spatial-Temporal Characteristics of Freeway Traffic Flows

碩士 === 中原大學 === 土木工程研究所 === 99 === Through the development of highway transportation, freeway has become an important transportation system in Taiwan. The large amounts of traffic flow make a serious impact on traffic environment and derive a variety of transportation problems. In order to present a...

Full description

Bibliographic Details
Main Authors: Ying-Yi Liu, 劉穎頤
Other Authors: Yu-Chun Liao
Format: Others
Language:zh-TW
Published: 2011
Online Access:http://ndltd.ncl.edu.tw/handle/65031859185637269751
Description
Summary:碩士 === 中原大學 === 土木工程研究所 === 99 === Through the development of highway transportation, freeway has become an important transportation system in Taiwan. The large amounts of traffic flow make a serious impact on traffic environment and derive a variety of transportation problems. In order to present a well-service highway traffic condition, it is necessary to use some strategies, such as ramp metering, to control travel space and time with optimal condition. The scheduling of traffic control system must be utilized by the characteristics of practical traffic. Therefore, the analysis of the transportation corridors provides an important reference value. The analysis of freeway traffic flow includes traffic volume, density, speed and other characteristics. Among them, the level of service (LOS) could be used as the evaluation criteria of transportation system. In order to determine LOS of freeway, this study applied fuzzy membership functions based on the reference of LOS in Highway Capacity Manual; furthermore, we segmented National Highway No.1 into various sections to present the mentioned characteristics, including LOS. The data of traffic volume was collected and organized from vehicle detectors set on National Highway No.1; we applied the cluster analysis of time and space segment respectively, and presented the results graphically. Moreover, we tested the result of clusters statistically to examine the optimality. Finally, we employed the results of cluster analysis for the strategies of transportation planning and ramp metering.