跳到主要內容

臺灣博碩士論文加值系統

(3.237.38.244) 您好!臺灣時間:2021/07/24 16:29
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:廖曉萍
研究生(外文):Liao, Hsiao-Ping
論文名稱:自小客車通勤者出發時間與路徑選擇動態行為之研究
論文名稱(外文):A Study of Car Drivers'' Commuting Behavior : Day-to-day Dynamic Departure Time and Route Choice
指導教授:董啟崇董啟崇引用關係
指導教授(外文):Chee-Chung Tong
學位類別:碩士
校院名稱:淡江大學
系所名稱:土木工程學系
學門:工程學門
學類:土木工程學類
論文種類:學術論文
論文出版年:1996
畢業學年度:84
語文別:中文
論文頁數:114
中文關鍵詞:動態行為模式出發時間與路徑選擇旅次鏈兩階段式問卷調查滿意規則無異帶
外文關鍵詞:Day-to-day Dynamic Behavior ModelDeparture Time and Route ChoiceTrip-chainTwo-stage SurveySatisficing RuleIndifference Band
相關次數:
  • 被引用被引用:10
  • 點閱點閱:213
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
  本研究旨在探討通勤者日復一日之出發時間與路徑選擇動態行為,及
旅次鏈的發生對兩決策行為之影響,並透過實際通勤資料之蒐集,以分析
通勤者受隨日變化之外在環境與內在認知的影響而產生之動態行為特性,
與影響決策行為之重要因素,並分別構建適當之決策變換行為模式,以解
釋不同型態之行為特性。  本研究採用兩階段式問卷調查方法,並以淡
水鎮至台北市通勤走廊之上班通勤者為實證對象。其方法為先以第一階段
敘述性偏好方式設計之問卷,取得靜態的通勤資料,以分析得整體一般行
為特性;再以第二階段日誌式問卷,取得個人連續數日之動態通勤資料,
以分析得個人之日復一日動態行為,最後,以該行為特性分析結果為依據
,應用多項羅吉特模式理論基礎及無異帶理念,構建出發時間與路徑變換
之動態行為與實證模式。  本研究方法與架構可作為未來相關研究之先
驗基礎,而研究成果則可彌補國內有關動態旅運行為理論之不足,並可提
供動態交通指派及交通控制相關研究一更近真實的行為假設基礎。本研究
所得之重要結果包括下列五點:1.通勤者之動態學習行為型態似具有〝週
期性〞,即其動態經驗累積過程以一週為始末,個人作決策僅以當週之通
勤經驗為參考,僅逢星期一會對過去相同日程之經驗有所注意。2.通勤者
有使用一主要路徑與在習慣時間範圍內出發之行為,故適於以〝滿意規則
〞機制描述其行為。3.影響出發時間決策無異帶之重要因素,包括個人決
策累積變換次數及路徑旅行時間標準差等變數,當上列變數值愈大,則其
無異帶將愈窄,即變換行為之頻率愈高。有關影響路徑決策無異帶之重要
因素,亦包括個人決策累積變換次數及路徑旅行時間標準差等變數,其中
累積變換次數變數之影響與出發時間維度相同,但路徑旅行時間標準差變
數僅於動態路徑基準下有顯著的影響,且當其值愈大,則無異帶將愈寬,
即變換行為之頻率愈低。4.例行性與非例行性旅次(鏈)對通勤決策行為
有不同的意義與影響,其中以非例行性旅次(鏈)所造成之出發時間與路
徑變換行為均較頻繁。5.本研究在比較多種不同變換基準定義之決策校估
無異帶模式結果,以〝動態10分鐘及眾數10分鐘〞兩種變換基準較適用於
出發時間無異帶模式,此外則以〝常用路徑〞變換基準較適用於路徑決策
無異帶模式。
The focus of this thesis is on the drivers'' day-to-day
dynamic departuretime and route choice behavior including the
influence of trip-chain as well.To realize commuters'' dynamic
behavior characteristic and the important factorsaffecting
travel behavior through analyzing actual commuting data.
Basedon these result, the dynamic switch behavior models were
established. A two-stage survey was employed to record such
behavior at Tamshui-Taipeicommuting corridor. In the first
stage, questionnaire with the stated-preferenceform was record
to capture the general characteristic. For the secondstage, a
travel diary was designed to record the day-to-day commuting
behaviorover a two-week period. Finally, the dynamic switch
behavior models were establishedthrough the construction of the
so-called "Indifference Band" concept. To reducethe complexity
of model estimation, a MNL model structure was then applied
tocalibrate model parameters. This study can provide useful
information as a priori for studying dynamictravel behavior for
both modelling structure and behavior analysis in the future.
Results of this study is also useful as supplement for issues
regarding dynamictraffic assignment and traffic control.
Hopefully, this study could be a benchmarkfor dynamic behavior
research in this country and call for more attentions inthis
regards for domestic researches. Major findings of this
thesis are summarized in the following :1.There exist some
"cyclic" phenomenon within a week for the day-to-daycommuting
choice behavior, where each Monday can be treated as a starting
dayfor a week-long day-to-day travel cycle.2.The application of
satisficing rule to construct day-to-day choice model seemsto be
appropreate, hence it is more adequate to describe the day-to-
day dynamiccommuting behavior as "switches" then "choices".3.
Factors significantly affecting the day-to-day departure time
switches are :the cummulative outcome failure and the standard
deviation of travel time,when the magnitude of these variables
would affect commuters switch departure timeindifference band in
the same direction; same would factors affecting day-to-day
routeswitch with the difference that the standard deviation of
travel time wouldpost negative effect on the route indifference
band.4.It exhibits stronger switching intensity of both
departure time and route switches fornon-routine trip-chain
trips.5.Among various definitions of "departure time switch" in
terms of switchthreshold showed that better parameter
estimation result was obtained when a10-minute threshold was
defined.While, definition of "route switch" in terms of "switch
from the main route" seemedto be most appropreate.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊