中小型鄉鎮接駁公車之路線設計方法

THE ROUTE DESIGN OF FEEDER BUS IN SMALL AND MEDIUM TOWNSHIP

在追求社會永續與大眾運輸無縫服務環境思維下，政府積極規劃接駁公車路線以填補大眾運輸縫隙，然中小型鄉鎮於交通分區上即為一交通分區，無法得知內部之運輸需求，故如何在規劃過程中，取得在地民眾運輸需求，以設計出符合民眾實際需求並能填補目前大眾運輸服務縫隙之接駁公車路線將是一項重要課題。雖然交通部公路總局在民國102年之研究中已針對臺灣偏鄉地區提出考量社區居民意識之大眾運輸路線規劃之方法，但該方法中之路線規劃分析仍屬於專家主觀判斷層面，缺乏一套完整之客觀分析方法。本研究結合運用越野競賽問題觀念，依據社區民眾所提出預設站點之重要性，提出在時間限制下能夠滿足最大服務需求接駁公車路線設計模式，並整合地理資訊系統與基因演算法發展兩階段之求解方法。經以新竹縣橫山鄉為應用案例，設計出以鄉公所為發車地點之4條接駁公車路線，不僅可提升14%之大眾運輸服務涵蓋家戶數，且該路線亦經過各主管機關代表之認可，顯示本研究所建立之路線設計方法可做為未來中小型鄉鎮接駁公車路線設計時之參考。

In order to increase the utilization of public transportation, the government actively planed feeder bus system to fill the gap of public transportation services. However, in the traditional transportation planning analysis, small and medium sized towns usually is a single traffic zone, so is unable to obtain internal transportation demand. Therefore, how to get the transport demand in the planning process and designed feasible feeder bus routes to satisfy actual demand and fill the gap of public transport services will be an important issue. This study combined the concepts of orienteering problem, geographic information systems and genetic algorithms to design the shuttle bus route. Using the concept of orienteering problem control point (score), give each stop a score to represent different importance. Application of the spatial analysis functions in geographic information system to build data and analysis the public transportation accessibility. Use of genetic algorithms to solving bus route design problem, and each route must to satisfy the time limit and the maximum score. In the real test case, the proposed method could increase the 14% public transportation service accessibility of Hengshan Township in Hsinchu County.