臺灣博碩士論文加值系統

摘要
運量預測在城市交通管理中具有重要的應用價值其中起迄之預測更能有效瞭解乘客的移動模式進而有效地安排列車運行時間表和車輛配置，以匹配實際需求，然而傳統方法常常無法有效應對節假日、特殊活動或突發事件散場對乘運量的影響，這導致預測準確度不高。為解決這一問題，本研究提出了一種類神經網路預測模型，結合活動標籤和歷史分時起迄數據，以提高對特殊活動期間乘客起迄分佈的準確預測能力。
該模型利用PyTorch框架中的nn.Module構建，輸入包括時段、進站點、出站點、人次數據、工作日或假日、異常事件標記、活動類型及起迄點間行車時間等特徵，輸出則為各站點的出站比例，同時本研究因起迄比例為機率分佈故使用交叉熵損失函數能有效提高模型對捷運站點起迄分佈的準確預測。
在學習率算法上，本研究從Adam轉為SGD，並手動調整學習率以避免模型輸出極端值，顯著改善了模型對高維度數據的適應性。此外本研究提出了針對臺北捷運的預測性調度和增強服務建議，包括找出與發生特殊活動站點關聯較強的迄點並提前於該站點做準備和制定活動合作策略，以提升整體運營效率和乘客滿意度。

Abstract
Passenger volume prediction in urban transportation management is crucial, particularly origin-destination (OD) prediction, which helps understand passenger movement patterns and efficiently schedule train operations and vehicle allocation. Traditional methods often fail to handle holidays, special events, or unexpected incidents, resulting in low prediction accuracy. To address this, we propose a neural network prediction model that combines activity labels with historical time-series OD data, improving prediction accuracy during special events.Our model, constructed using the nn.Module in PyTorch, takes inputs such as time periods, entry and exit stations, passenger counts, workdays or holidays, anomaly event markers, activity types, and travel times between OD points. The output is the exit proportion at each station. Given that the OD proportion is a probability distribution, we use the cross-entropy loss function to enhance the model's accuracy in predicting OD distributions for metro stations. For the learning rate algorithm, we switch from Adam to SGD and manually adjust the learning rate to avoid extreme values in the model output, significantly improving the model's adaptability to high-dimensional data. Additionally, we propose predictive scheduling and enhanced service recommendations for the Taipei Metro, including preemptively adjusting train intervals, real-time updates of passenger flow information, and developing event cooperation strategies to boost overall operational efficiency and passenger satisfaction. Ultimately, we use ticket OD data to calculate inter-station volume.

目錄
摘要 i
Abstract ii
誌謝 iii
目錄 iv
圖目錄 vi
表目錄 viii
第一章 緒論 1
第二章 文獻回顧 3
2-1機器學習於交通變量預測之應用 3
2-2捷運運量預測變數選擇 4
2-3異常值判斷 5
第三章 研究方法 7
3-1 Z分數對異常值的篩選 7
3-2類神經網路(ANNs)的起迄分佈預測方法 8
3-3利用交叉熵衡量模型 9
3-4模型之評估指標 11
第四章 模型訓練與驗證 12
4-1資料前處理之過程 12
表4行政院全國行事曆 17
4-2以臺北捷運板橋站為例類神經網路訓練過程 23
4-2.1類神經網路架構 23
4-2.2模型優化算法設置 23
4-3結果驗證 34
第五章 結果分析 37
5-1板橋、大安森林公園、中正紀念堂、臺北101/世貿活動散場起迄分佈 37
5-2淡水、臺北小巨蛋站活動散場起迄分佈 42
第六章 結論與建議 46
6-1結論 46
6-2建議 47
參考文獻 48

參考文獻
[1] 林子鈞，2020，以深度學習方法預測大型活動對臺北捷運運量之影響，國立臺灣大學地理環境資源學系，碩士論文，臺北市
[2] 翁宇鴻，2022，應用手機信令預測捷運站間量之研究，國立中央大學土木工程學系，碩士論文，桃園市。
[3] 陳惠國，2022，研究分析方法講義，國立中央大學土木工程學系，桃園市。
[4] Liu, L., Zhu, Y., Li, G., Wu, Z., Bai, L., and Lin, L., 2022, Online metro origin-destination prediction via heterogeneous information aggregation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(3), 3574-3589.
[5] Vanajakshi, L., and Laurence R., 2007, Support vector machine technique for the short term prediction of travel time. IEEE Intelligent Vehicles Symposium, 600-605.
[6] Jeong, Y. S., Byon, Y. J., Manoel C. N., and Easa., 2013, Supervised weighting-online learning algorithm for short-term traffic flow prediction. IEEE Transactions on Intelligent Transportation Systems , 1700-1707.
[7] Jiber, M., Lamouik, I., Ali, Y., and Sabri, M. A., 2018, Traffic flow prediction using neural network. In 2018 International Conference on Intelligent Systems and Computer Vision (ISCV) 1-4.
[8] Gallo, M., De Luca, G., D’Acierno, L., and Botte, M., 2019, Artificial neural networks for forecasting passenger flows on metro lines. Sensors, 19(15), 3424.
[9] Kusonkhum, W., Srinavin, K., Leungbootnak, N., and Chaitongrat, T., 2022, Using a Machine Learning Approach to Predict the Thailand Underground Train’s Passenger. Journal of Advanced Transportation.
[10] Hou, Y., and Edara, P., 2018, Network scale travel time prediction using deep learning. Transportation Research Record, 2672(45), 115-123.
[11] Jia, Y., Wu, J.,and Xu, M., 2017.,Traffic flow prediction with rainfall impact using a deep learning method. Journal of Advanced Transportation 2017(722), 1-10
[12] Liu, L., and Chen, R. C., 2017, A MRT daily passenger flow prediction model with different combinations of influential factors. In 2017 31st International Conference on Advanced Information Networking and Applications Workshops (WAINA) 601-605.
[13] Shiao, Y. C., Liu, L., Zhao, Q., and Chen, R. C., 2017, Predicting passenger flow using different influence factors for Taipei MRT system. In 2017 IEEE 8th International Conference on Awareness Science and Technology (iCAST) , 447-451.
[14] Wu, J., and Liao, H., 2020, Weather, travel mode choice, and impacts on subway ridership in Beijing. Transportation research part A: policy and practice, 135, 264-279.
[15] Verma, T., Sirenko, M., Kornecki, I., Cunningham, S., & Araújo, N. A. , 2021, Extracting spatiotemporal commuting patterns from public transit data. Journal of Urban Mobility, 1, 100004.
[16] Mondal, M. A., and Rehena, Z., 2020, Road traffic outlier detection technique based on linear regression. Procedia Computer Science, 171, 2547-2555.
[17] Kim, S., Shibuya, T., Toride, S., and Endo, Y., 2022, A Human-Flow Analysis Based on PCA: A Case Study on Population Data Near Railway. In 2022 2nd International Conference on Robotics, Automation and Artificial Intelligence (RAAI).
[18] Reddi, S. J., Kale, S., and Kumar, S., 2019, On the convergence of adam and beyond. arXiv preprint arXiv:1904.09237.