標題: 智慧多模式交通系統規劃與動態派遣模式
Long-term Operation Programming and Dynamic Dispatching Operation Modeling for Intelligent Multi-mode Transit System
作者: 吳政儒
Jeng-Ru Wu
許鉅秉
Jiuh-Biing Sheu
運輸與物流管理學系
關鍵字: 多目標規劃法;動態發車間距;模糊數學規劃法;Dynamic headways;Intelligent Multi-mode Transit System;Multi-objective programming
公開日期: 2005
摘要: 本研究探討日本豐田汽車公司最新研發之一種兼有軌道系統和道路系統優點的新型中距離、中量運輸系統IMTS(Intelligent Multi-mode Transit System),針對IMTS可在專用道上排成隊列自動行駛、可根據需要隨時改變車隊(platoon)的車輛數、亦可行駛於一般市區道路上下乘客(節省換車手續及時間)、及在一般公車設備工廠即可維修(節省系統建設與維護成本)等優點,進行營運模式構建。 本研究提出之模式架構包含了兩個模組,第一個是IMTS營運計劃多目標規劃模式之建立,同時考慮營運者與使用者的立場,以營運者每日長期營運總成本最小化以及使用者每日總旅行時間最小化為目標,決定出發車班距、營運所需車輛數以及路線站距。第二個為動態即時派遣模組,此模組可計算出每個決策點各繞徑路線之車輛成本增量,並與成本增量門檻值作比較,建立一評選機制,以挑選當前最迫切需要服務之派遣路線,推估動態發車間距,進行整合性的最適營運控制模式分析並與現有捷運系統及接駁公車營運績效作比較。 本研究以臺北捷運南港線與周圍接駁公車系統的整合路網進行實例應用,利用模糊數學規劃法透過目標達成滿意度之概念化方式進行求解。研究結果顯示:IMTS發揮多模運具優勢,較現行捷運系統搭配接駁公車的營運方式,大幅減少乘客平均旅行時間。且以「動態派車」方式進行營運,在中運量的旅客需求型態下,優於傳統「固定頻率」發車方式,可有效節省整體系統成本,使派車機制與管理系統更為完善。
Determining dynamic headways for correctly identifying passenger demands and quickly responding to those needs with Intelligent Multi-mode Transit System (IMTS) operations control strategies is vital to the development of advanced public transportation systems in urban areas. The proposed model primarily involves two levels of functionality:(1) long-term operation programming utilizing multi-objective programming methods, and (2) identification of service strategies coupled with the associated bus service segments using dynamic dispatching operation model in response to variances in passenger demand attributes. The numerical results significantly reveal the potential advantages of the proposed methodology. Particularly, the travel time can be reduced by 51.6%, compared to the existing operational performance in the specific case studied. We expect that this study can make available the proposed method with benefits not only for programming IMTS long-term operations, but also for deciding dynamic dispatching strategies in a comprehensive extent.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT009336517
http://hdl.handle.net/11536/79611
Appears in Collections:Thesis


Files in This Item:

  1. 651701.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.