以行動信令資料推估觀光需求為基礎之公車服務最佳化模式

Bus Service Optimization Model based on the Tourist Demand Estimated by Cellular Data

王以萱(Yi-Hsuan Wang)

指導教授 ： 邱裕鈞

國立交通大學/管理學院/運輸與物流管理學系/碩士(2018年)

摘要

旅運行為之研究可以使交通主管機關更有效地做運輸規劃及城市管理，現有資料來源包含旅運調查資料、ETC、固定式車輛偵測器、電子票證、GPS、行動網路信令資料，這些資料都有助於交通管理，其中行動網路信令資料具有高覆蓋率、更新頻率迅速、大樣本、以及極低的成本的優點，可以進行更深入的個體行為研究，對規劃大眾運輸系統十分有幫助。

公車服務設計（排程與排班）大多係基於起迄旅次需求推估，而非旅次鏈需求，但用路人在旅程中不太可能在第一個旅次選擇搭公車，但卻在接下來的旅次卻選擇開車，足見同一旅次鏈之各旅次行為決策，必定彼此互相影響，這是以旅次鏈為基礎建立公車服務設計的重要性。基此，本研究利用行動信令資料所推估而得的旅次鏈資料為基礎，建立一個數學規劃模式，以羅吉特模式推估使用者搭乘公車路線的機率，考量總成本限制，設計使用者總搭乘人數最大化下之公車服務路網及行駛班距，公車路線包含一般路線及循環路線，為驗證所提模式之可用性，進行簡例驗證，驗證結果證實對於只停留一個景點的旅次鏈，傾向設計一般路線，對於停留兩點以上的旅次鏈，傾向設計循環路線，且可以反應不同時段之需求設計分時之班距。

為進一步證實模式的實用性，以宜蘭為實例驗證地區，為推估實際需求，本研究依據觀光局國內主要觀光遊憩據點之月統計資料推估放大因子，其值為1.82。旅次鏈計算結果顯示，大多旅次鏈需求係往返轉運站及兩個觀光景點，五峰旗瀑布與國立傳統藝術中心，因此，模式設計出4條公車路線（2條對開線及2條循環線）以銜接主要觀光景點及轉運站，以載運最多旅客，每條路線都依據旅次鏈需求有不同的班次設計。

關鍵字：旅次鏈、信令資料、公車路線、公車頻率、旅遊景點

關鍵字

旅次鏈 ； 信令資料 ； 公車路線 ； 公車頻率 ； 旅遊景點

並列摘要

Travel behaviors have been studied for decades to guide transportation development and management. There are many sources of travel behavior data, including traditional travel survey, ETC data, roadside detector, electronic tickets on MRT and city bus, GPS, and cellular data. Comparing to other data source, cellular data have advantages of high coverage, high updating frequency, large sample size and nearly zero additional cost, therefore cellular data can be used to capture and estimate travel behaviors which is helpful for transportation system planning.

Bus service designs (routing and scheduling) are mainly based on estimated trip demand ignoring trip chaining behaviors. However, it is unlikely for a traveler to take a bus for the first trip in a journey, but decide to drive a car in the following trips. That is, the decisions of trips of a traveler are highly intertwined, explaining the importance of activity based demand modeling. Based on this, this study aims to propose a mathematical programming model to optimize bus service based on trip-chaining demand estimated by cellular data. The model estimates the probability of choosing bus by logit model based on trip chain data. The objective function is to maximize number of choosing bus under total cost constraint. The model contains end-to-end bus routes and circular bus routes. To validate the proposed model, a case study on an exemplified example is conducted. It is found that travelers visiting only one spot tend to choose end-to-end bus routes whereas those who visit more than one spot tend to choose circular bus routes. Additionally, the proposed model can simultaneously determine various headways according to dynamically varying travel demand.

To further demonstrate the applicability of the proposed model, a case study on Yilan County is conducted. To estimate the real trip chaining demand, the cellular trip chaining data have been enlarged by 1.82 times according to the number of tourist statistics of major tourist attractions. The result shows that majority of the trip chains are traveling among transfer stations and two major attractions: Wuqifeng Waterfall and National Center for Traditional Arts in Yilan. A total of four bus routes are recommended, two of them are end-to-end and others are circular routes, all of them connecting top visited tourist attractions and transfer stations. Each of bus routes has various bus frequencies depending upon the dynamic demand.

Keywords: Trip chaining, cellular data, bus route, bus frequency, tourist attractions.

並列關鍵字

Trip chaining ； cellular data ； bus route ； bus frequency ； tourist attractions

參考文獻

林育瑄(2013)。模化旅運者運具選擇行為之動態性與異質性。交通大學運輸與物流管理學系學位論文, 1-72。

Aguiléra, V., Allio, S., Benezech, V., Combes, F., & Milion, C. (2014). Using cell phone data to measure quality of service and passenger flows of Paris transit system. Transportation Research Part C: Emerging Technologies, 43, 198-211.

Alexander, L., Jiang, S., Murga, M., & González, M. C. (2015). Origin–destination trips by purpose and time of day inferred from mobile phone data. Transportation Research Part C: Emerging Technologies, 58, 240-250.

Berlingerio, M., Calabrese, F., Di Lorenzo, G., Nair, R., Pinelli, F., & Sbodio, M. L. (2013, September). AllAboard: a system for exploring urban mobility and optimizing public transport using cellphone data. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases (pp. 663-666). Springer, Berlin, Heidelberg.

Demissie, M. G., Phithakkitnukoon, S., Sukhvibul, T., Antunes, F., Gomes, R., & Bento, C. (2016). Inferring passenger travel demand to improve urban mobility in developing countries using cell phone data: A case study of Senegal. IEEE Transactions on Intelligent Transportation Systems, 17(9), 2466-2478.