交通問題是都市化帶來的嚴峻考驗之一。而發展公共運輸，提升其使用率是解決交通問題的一種有效方式。公共汽車作為最常見的公共運輸工具之一，在整個公共運輸系統中發揮著非常重要的作用。因此，如何提高公車的服務質量，讓人們更願意搭乘公車出行，以此提升公車的使用率，是一個非常重要的問題。本文提出了一種由多個子模型組合的公車運行預測模型。通過該模型可達成僅調整公車的發車時間，便提高公車服務質量的目的：降低擁擠程度，縮短通行時間。具體來說，我們首先結合公車票卡記錄資料和其他公開資料，如天氣狀況、交通事故等，並利用深度學習的方法構建多個子模型，預估每一個公車站點的上下車人數，公車在站點的停留時間，以及站點之間的運輸時間。通過子模型之間的組合，估計未來公車的整體運行狀況。因此，我們可以在公車的發車間隔內，選擇的最佳發車時間來優化整體的運輸時間和擁擠程度，即達到最佳的服務質量。最後，對台中市300路公車實際資料的實驗結果表明，我們的公車發車時間調整方法對提升其服務質量是有效的。

The traffic problem is one of the most serious problems brought about by urbanization. The development of public transport and the increase in its usage rate are an effective way to solve the traffic problem. As one of the most common forms of public transport, buses play an important role in the entire public transportation system. Therefore, how to improve the service quality of the buses and make people more willing to take buses to increase their usage rate is a very critical issue. Here we consider improving service quality as to lower the crowdedness degree and to lessen the transfer time. To this end, in this thesis, we purposed a bus operation prediction model with multiple sub-model combinations. By only dynamically adjusting the departure time of the bus, it can achieve the purpose of improving the service quality of the bus. Specifically, we first combined buses fare card records data and open data, such as weather conditions and traffic accidents, use the deep learning method to build multiple sub-models, estimate the number of passengers who board and alight the bus, boarding and alighting time and running time. Through the combination of sub-models, estimate the overall operation of the future bus. Therefore, we can choose the best departure time within the bus take-off interval to optimize the overall transfer time and the crowdedness degree of the bus, which is the best service quality. Finally, experimental results on real-world data of Taichung City bus route #300 show that our bus departure time adjustment method is effective for improving its service quality.

[1] Rodrigue, J. P., Comtois, C., & Slack, B. (2016). The geography of transport systems. Routledge.
[2] Farahani, R. Z., Miandoabchi, E., Szeto, W. Y., & Rashidi, H. (2013). A review of urban transportation network design problems. European Journal of Operational Research, 229(2), 281-302.
[3] The Ministry of Transportation and Communications of the Republic of China (2017). Abstract analysis of the survey on the daily use of vehicles in 2016.
[4] Balcombe, R., Mackett, R., Paulley, N., Preston, J., Shires, J., Titheridge, H., ... & White, P. (2004). The demand for public transport: a practical guide.
[5] Zheng, Y. (2015). Trajectory data mining: an overview. ACM Transactions on Intelligent Systems and Technology (TIST), 6(3), 29.
[6] Munizaga, M. A., & Palma, C. (2012). Estimation of a disaggregate multimodal public transport Origin–Destination matrix from passive smartcard data from Santiago, Chile. Transportation Research Part C: Emerging Technologies, 24, 9-18.
[7] Trépanier, M., Tranchant, N., & Chapleau, R. (2007). Individual trip destination estimation in a transit smart card automated fare collection system. Journal of Intelligent Transportation Systems, 11(1), 1-14.
[8] Lee, S. G., & Hickman, M. (2014). Trip purpose inference using automated fare collection data. Public Transport, 6(1-2), 1-20.
[9] Trepanier, M., & Chapleau, R. (2006). Destination estimation from public transport smartcard data. IFAC Proceedings Volumes, 39(3), 393-398.
[10] Nunes, A. A., Dias, T. G., & e Cunha, J. F. (2016). Passenger journey destination estimation from automated fare collection system data using spatial validation. IEEE transactions on intelligent transportation systems, 17(1), 133-142.
[11] Ma, X., Wu, Y. J., Wang, Y., Chen, F., & Liu, J. (2013). Mining smart card data for transit riders’ travel patterns. Transportation Research Part C: Emerging Technologies, 36, 1-12.
[12] Briand, A. S., Côme, E., Trépanier, M., & Oukhellou, L. (2017). Analyzing year-to-year changes in public transport passenger behaviour using smart card data. Transportation Research Part C: Emerging Technologies, 79, 274-289.
[13] van Oort, N., Drost, M., Brands, T., & Yap, M. (2015, July). Data-driven public transport ridership prediction approach including comfort aspects. In Conference on Advanced Systems in Public Transport, Rotterdam, The Netherlands.
[14] Bagchi, M., & White, P. (2004). What role for smart-card data from bus systems?. Municipal Engineer, 157(1), 39-46.
[15] Jang, W. (2010). Travel time and transfer analysis using transit smart card data. Transportation Research Record: Journal of the Transportation Research Board, (2144), 142-149.
[16] Smart, M., Miller, M. A., & Taylor, B. D. (2009). Transit stops and stations: transit managers’ perspectives on evaluating performance. Journal of Public Transportation, 12(1), 4.
[17] Gschwender, A., Munizaga, M., & Simonetti, C. (2016). Using smart card and GPS data for policy and planning: The case of Transantiago. Research in Transportation Economics, 59, 242-249.
[18] Ceapa, I., Smith, C., & Capra, L. (2012, August). Avoiding the crowds: understanding tube station congestion patterns from trip data. In Proceedings of the ACM SIGKDD international workshop on urban computing (pp. 134-141). ACM.
[19] Karlaftis, M. G., & Vlahogianni, E. I. (2011). Statistical methods versus neural networks in transportation research: Differences, similarities and some insights. Transportation Research Part C: Emerging Technologies, 19(3), 387-399.
[20] Yu, J., & Yang, X. G. (2006). Estimating a Transit Route OD Matrix from On-Off Data through an Artificial Neural Network Method. In Applications of Advanced Technology in Transportation (pp. 467-472).
[21] Lin, Y., Yang, X., Zou, N., & Jia, L. (2013). Real-time bus arrival time prediction: Case study for Jinan, China. Journal of transportation engineering, 139(11), 1133-1140.
[22] Zhang, J., Zheng, Y., & Qi, D. (2017, February). Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Prediction. In AAAI (pp. 1655-1661).
[23] Polson, N. G., & Sokolov, V. O. (2017). Deep learning for short-term traffic flow prediction. Transportation Research Part C: Emerging Technologies, 79, 1-17.
[24] Nam, D., Kim, H., Cho, J., & Jayakrishnan, R. (2017, January). A Model Based on Deep Learning for Predicting Travel Mode Choice. In Proceedings of the Transportation Research Board 96th Annual Meeting Transportation Research Board, Washington, DC, USA (pp. 8-12).
[25] Jung, J., & Sohn, K. (2017). Deep-learning architecture to forecast destinations of bus passengers from entry-only smart-card data. IET Intelligent Transport Systems, 11(6), 334-339.
[26] Transportation Research Board (TRB). (2003). Transit Capacity and Quality of Service Manual, 2nd Edition, Transportation Research Board, USA.
[27] Chakroborty, P., & Kikuchi, S. (2004). Using bus travel time data to estimate travel times on urban corridors. Transportation Research Record: Journal of the Transportation Research Board, (1870), 18-25.
[28] Robinson, S., Narayanan, B., Toh, N., & Pereira, F. (2014). Methods for pre-processing smartcard data to improve data quality. Transportation Research Part C: Emerging Technologies, 49, 43-58.
[29] LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324.
[30] Ma, X., Dai, Z., He, Z., Ma, J., Wang, Y., & Wang, Y. (2017). Learning traffic as images: a deep convolutional neural network for large-scale transportation network speed prediction. Sensors, 17(4), 818.