公車路線複雜度與路線圖設計對公車路線選擇行為之影響

Abstract

臺北市有三百多條公車路線，尤其大型的公車場站擁有眾多公車路線，對於不熟悉的公車使用者，要找到可以搭乘的路線實屬困難。因此本研究的目的有三個：第一，找出使用者注重哪些影響路線選擇複雜度的因素；第二，衡量使用者對於不同的公車路線圖資訊的提供，是否提升使用者對於公車路線的認知能力；第三，提出大型公車場站路線地圖的設計建議。本研究分為兩階段，先使用現有的公車路線圖進行實驗，找出影響用路人路線選擇複雜度的因素；接著依據找到的因素與參考其他城市公車路線圖與相關文獻，設計出新的公車路線圖，試著改善目前地圖的設計方式、降低乘客在路線選擇的複雜度。 本研究以捷運公館站之公車站為研究對象，第一階段實驗使用行經該站的54條路線作為情境的設計，以5個因子將路線分類，包含路線數、公車行駛距離、班次間距變異係數、剩餘站數變異係數、繞路程度比例，除了繞路程度分為兩種水準值，其餘4個因子皆分為低中高三種水準值，並透過直交表的實驗組合選出15個情境，每3個情境作1次複雜度的排序，共得到161份樣本，並使用等級次序羅吉特進行分析。結果發現，路線數為主要造成選擇路線的複雜因子，公車行駛距離為其次，班次間距、剩餘站數、繞路程度對於路線選擇的複雜程度影響則相對薄弱。 第二階段以改善路線數造成的複雜度為主，輔以蛛網圖概念，設計出一簡化的公車路線圖，具有保留路線與站位的相對位置、以單一線段表示多條路線經過、僅畫出方圓4公里之路線、有街道與公車站牌搭乘位置、以顏色區分不同行駛方向等特性。並發放問卷讓受測者使用新的公車路線圖，共收得87份問卷，以羅吉特迴歸模式分析。結果顯示本研究所設計的公車路線圖比現有的公車路線圖較好。

Taipei city has more than 300 bus routes, especially giant bus stations serves many bus routes. It is difficult for unfamiliar bus users to find a route that can be taken. There are three purposes of this study: First, find out the factors that affect bus network complexity for the bus users; second, we use of information on different map of bus routes, whether to enhance the user's ability to understand the bus route; third, we propose a recommendation of designing giant bus station route map. The study is divided into two stages. First, the existing bus route map is used to find out the factors that affect the bus network complexity. Then, according to the factors found, the reference to other urban bus routes map and related literature. Try to improve the design of the current bus map and reduce the bus network complexity. In this study, the object of study of the bus station is MRT Gongguan station. The first stage experiment uses the 54 routes of the station as the design of the situation. The route is classified by five factors, including the number of routes, the distance of the bus, shift interval coefficient of variation, number of remaining stops, and percentage of detours. In addition to the percentage of detour is divided into two levels, the other four factors are divided into low, medium and high level. Through the orthogonal table of the experimental combination to select 15 cases, each three cases were ranked the complexity, and 161 samples were obtained and analyzed using the Rank-ordered logit. It is found that the number of routes is mainly factor affect the bus network complexity, the distance of the bus is the minor factor, shift interval coefficient of variation, number of remaining stops, percentage of detours are relatively weak for the bus network complexity. In the second stage, this research mainly changed the complexity caused by number of routes, and supplemented by the concept of the spider map to design a simplified bus route map. It has the relative position of the route and the station, and use single segment represent nultiple routes. We only draw radius of 4kms of the map. There are sign of streets and bus station. Using color to distinguish the different directions of routes. And issued a questionnaire for the subjects to use the new bus map, received a total of 87 questionnaires, and used Logit model to analysis. The results show that the bus map designed in this study is better than the existing bus map.