簡易檢索 / 詳目顯示

研究生: 陳弘毅
Hong-Yi Chen
論文名稱: 於共享單車系統中採用威布爾分布的單車數量預測
Using the Weibull Distribution to Predict the Number of Bikes in a Bike-Sharing System
指導教授: 賴源正
Yuan-Zheng Lai
口試委員: 楊傳凱
Chuan-Kai Yang
陳彥宏
Yan-Hong Chen
學位類別: 碩士
Master
系所名稱: 管理學院 - 資訊管理系
Department of Information Management
論文出版年: 2021
畢業學年度: 109
語文別: 中文
論文頁數: 41
中文關鍵詞: 站點分類共享單車威布爾分布
外文關鍵詞: station classification, shared bicycle prediction, Weibull distribution
相關次數: 點閱:351下載:0
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • 隨著近年共享經濟的蓬勃發展,共享單車已經成為隨處可見的便利公共代步工具。目前單車數量的站點信息僅能提供目前該站點現有單車數量,但難以預測使用者到達該站時是否滿站/空站,從而造成借還車時等待時間的浪費。因此我們提出了一種基於威布爾分布的方法來預測單車未來短暫時刻的數量。我們針對共享單車在工作日使用習慣和特點,提出了基於空間上站點分類和時間上威布爾分布的共享單車站內數量的預測方法。此方法為使用預處理好的共享單車站點歷史數據,對站點單車的增量活躍區/減量活躍區進行劃分,再採用威布爾分布擬合活躍區的機率密度函數,因此可透過當前共享單車數量和機率密度函數,計算得出共享單車的預測數量。本方法並使用了全天預測結果的R-Square值進行評估,表現最佳站點之R-Square值為0.97134,次佳站點的R-Square值亦有0.83617,證實本方法的預測的準確性。


    With the rapid development of the sharing economy in recent years, shared bicycles have become a convenient transportation tool everywhere. Currently the station can provide only the number of existing bicycles at the station, but this information can not be used to predict whether the station is full/empty when the user arrives. This might result in a long waiting time when users borrow or return the bicycles. Therefore, we propose a prediction method based on Weibull distribution to predict the number of bicycles in the short-term future. Aiming at the usage habits and characteristics of shared bicycles during workdays, this prediction method for the number of shared bicycles is based on the classification of stations in space and Weibull distribution in time. This method uses the pre-processed historical data of shared bicycle sites to divide the incremental active area/decrease active area of the bicycles on the site, and then uses the Weibull distribution to fit the probability density function of the active area, so it can be shared through the current sharing The number of bicycles and the probability density function are calculated to obtain the predicted number of shared bicycles. The R-Square value of all-day forecast results is used for the method evaluation. Among all investigated stations, the best R-Square value is 0.97134 and the second one is 0.83617, which verify the prediction accuracy of our proposed method.

    目錄 摘要 I Abstract II 目錄 III 圖目錄 V 表目錄 VI 第壹章 簡介 1 第貳章 背景 3 一、 相關文獻 3 第参章 威布爾分布 5 一、 威布爾趨近法 5 二、 選擇威布爾分布的原因 5 第肆章 實驗方法 9 一、 威布爾分布預測共享單車流程圖 9 二、 數據集介紹 10 三、 數據預處理 11 四、 站點類型劃分 14 五、 增量活躍區、減量活躍區、平緩區劃分 19 六、 共享單車數量預測 20 第伍章 實驗結果 23 一、 預設參數值 23 二、 R-Square值 23 三、 殘差圖 24 第陸章 結論與展望 31 參考文獻 32 圖目錄 圖 31 威布爾累計曲線擬合活躍上升區單車數量變化 6 圖 41 威布爾分布預測共享單車流程圖 9 圖 42 篩選異常點處理 11 圖 43 補充中值處理 12 圖 44 波峰波谷計算及分區處理 13 圖 45 A類邊緣辦公區附近站點 15 圖 46 B類邊緣住宅區附近站點 16 圖 47 C類上下段辦公區附近站點 17 圖 48 D類上下段辦公區附近站點 18 圖 49 E類上下晚三段式辦公學習區附近站點 18 圖 410 F類通勤繁雜的捷運站公園等附近站點 19 圖 5-1 各站點類型R-Square的散點圖 26 圖 5-2 A型殘差圖 26 圖 5-3 B型殘差圖 26 圖 5-4 C型殘差圖 27 圖 5-5 D型殘差圖 28 圖 5-6 E型殘差圖 28 圖 5-7 F型殘差圖 29 表目錄 表 2-1 現有的共享單車流量預測方法對比 4 表 3-1 威布爾累積分布表 8 表 4-1 車站數據形式範例 11 表 4-2 共享單車站點劃分表 15 表 4-3 符號表 20

    [1]K. Su, J. Li, and H. Fu, "Smart city and the applications," in Proc. IEEE Int. Conf. Electron. Commun. Control (ICECC), Sept. 2011.
    [2]J. J. Yang,B. Z. Guo,Z. H. Wang, and Y. L. Ma,"Hierarchical Prediction Based on Network-Representation-Learning-Enhanced Clustering for Bike-Sharing System in Smart City" IEEE Internet of Things Journal, vol. 8, no. 8, pp. 6416-6424, April. 2021.
    [3]A. R. Al-Ali, Raafat Aburukba, A. H. Riaz, Ahmad Al Nabulsi, D. Khan, S. Khan and M. Amer, " IoT-Based Shared Community Transportation System Using e-Bikes," 2021 5th International Conference on Smart Grid and Smart Cities (ICSGSC), June 18-20, 2021.
    [4]J. F. Li, C. R. Ren, B. Shao, Q. H. Wang, M. He, and J. Dong, " A Solution For Reallocating Public Bike Among Bike Stations," Supply Chain Management and Logistics Research, April 11-14 , 2012.
    [5]M. Dell’Amico, M. Iori, S. Novellani, and A. Subramanian, ‘‘The bike sharing rebalancing problem with stochastic demands,’’ Transp. Res. B, Methodol, vol. 118, pp. 362-380, Dec. 2018.
    [6]M. J. He, X. W. Ma, and Y. C. Jin, ‘‘Station Importance Evaluation in Dynamic Bike-Sharing Rebalancing Optimization Using an Entropy-Based TOPSIS Approach,’’ IEEE Access, vol. 9, pp. 38119-38131, March.2021.
    [7]S. W. Huang, ‘‘台灣人為什麼不喜歡無樁共享單車?|越界華文答問,’’ 2018. Retrieved from https://zhuanlan.zhihu.com/p/40615477
    [8]Romain Giot, and Raphaël Cherrier, "Predicting Bikeshare System Usage Up to One Day Ahead," 2014 IEEE Symposium on Computational Intelligence in Vehicles and Transportation Systems (CIVTS), Dec 9-12, 2014.
    [9]F. H. Huang, S. J. Qiao, J. Peng, and B. Guo, ‘‘A Bimodal Gaussian Inhomogeneous Poisson Algorithm for Bike Number Prediction in a Bike-Sharing System,’’ IEEE Transactions on Intelligent Transportation Systems, vol. 20, no. 8, pp. 2848-2857, August. 2019.
    [10]Y. Li, Y. Zheng, H. Zhang, and L. Chen, "Traffic Prediction in a Bikesharing System," Proceedings of the 23rd SIGSPATIAL International Conference on Advances in Geographic Information Systems,pp. 1-10,Nov.2015.
    [11]Di Chai,Leye Wang, and Qiang Yang, "Bike Flow Prediction with Multi-Graph Convolutional Networks," Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, Nov. 2018.
    [12]M. Mohsin, and K.V.S. Rao, " Estimation of Weibull Distribution Parameters and Wind Power Density for Wind Farm Site at Akal at Jaisalmer in Rajasthan," 2018 3rd International Innovative Applications of Computational Intelligence on Power, Energy and Controls with their Impact on Humanity (CIPECH), Nov. 2018.
    [13]L. Hribar, and D. Duka, "Weibull distribution in modeling component faults," Proceedings ELMAR-2010, Sept. 2010.
    [14]S. Reissa, and K. Bogenbergera, "A Relocation Strategy for Munich's Bike Sharing System: Combining an operator-based and a user-based Scheme," Transportation Research Procedia, vol. 22, 2017, pp. 105-114.
    [15]Ramkumar Harikrishnakumar, Sima E. Borujeni, Alok Dand, and Saideep Nannapaneni, " A Quantum Bayesian Approach for Bike Sharing Demand Prediction," 2020 IEEE International Conference on Big Data, Dec 10-13, 2020.
    [16]Romain Giot, and Raphaël Cherrier, "Predicting Bikeshare System Usage Up to One Day Ahead," 2014 IEEE Symposium on Computational Intelligence in Vehicles and Transportation Systems (CIVTS), Dec 9-12, 2014.
    [17]Kyoungok Kim, "Spatial Contiguity-Constrained Hierarchical Clustering for Traffic Prediction in Bike Sharing Systems," IEEE Transactions on Intelligent Transportation Systems ( Early Access ), Feb 17, 2021.
    [18]Ananta Kumar Das, Amogh Manoj Joshi, and Subhasish Dhal, "A Machine Learning Based Bike Recommendation System Catering To User’s Travel Needs," 2020 IEEE 17th India Council International Conference (INDICON), Dec 10-13, 2020.
    [19]Naroa Coretti Sanchez, Luis Alonso Pastor, and Kent Larson, "Autonomous Bicycles: A New Approach To Bicycle-Sharing Systems," 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), Dec 20-23 , Sept. 2020.
    [20]Naroa Coretti Sanchez, Luis Alonso Pastor, and Kent Larson, "Autonomous Bicycles: A New Approach To Bicycle-Sharing Systems," 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), Sept.2020.
    [21]Haitao Xu, and Feng Duan, and Pan Pu, "Dynamic bicycle scheduling problem based on short-term demand prediction," Applied Intelligence, vol.49, pp.1968-1981, Dec.2018.
    [22]Jian Jiang, Fei Lin, Jin Fan, Hang Lv, and Jia Wu, "A Destination Prediction Network Based on Spatiotemporal Data for Bike-Sharing," Hindawi Complexity, vol. 2019, Jan. 2019.

    無法下載圖示 全文公開日期 2024/10/21 (校內網路)
    全文公開日期 本全文未授權公開 (校外網路)
    全文公開日期 本全文未授權公開 (國家圖書館:臺灣博碩士論文系統)
    QR CODE