研究生: |
何旺宗 He, Wong-Zone |
---|---|
論文名稱: |
資料融合技術結合類神經網路對高速公路事件延遲時間預測之研究 Data Fusion with Artificial Neural Network in Accident Duration Time Forecast on Freeway |
指導教授: |
魏健宏
Wei, Chien-Hung |
學位類別: |
碩士 Master |
系所名稱: |
管理學院 - 交通管理科學系 Department of Transportation and Communication Management Science |
論文出版年: | 2010 |
畢業學年度: | 99 |
語文別: | 中文 |
論文頁數: | 90 |
中文關鍵詞: | 事件延遲 、時間預測 、平均絕對誤差比率 、資料融合 、類神經網路 |
外文關鍵詞: | Accident Duration, Time Forecast, MAPE, Data Fusion, Artificial Neural Networks |
相關次數: | 點閱:133 下載:5 |
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 |
高速公路由於具有出入管制之性質,當事故發生時除了造成人員與車輛的損傷外,往往造成後方車輛的堵塞,因此事件持續時間的預測,可以提供後方駕駛人必要的交通資訊,讓駕駛人避開事件路段,減少高速公路因為事件產生之擁擠。受限於現有即時資訊系統建置的數量與區位,如何在有限的資訊記錄中準確預測出事件持續時間的長短,是本研究第一個面臨的問題。另外,該如何整合事件發生資訊與當時道路交通相關資訊來進行預測事件持續時間,在偵測器帶來多樣訊息的過程中,由於格式單位的不一致性,造成資料判斷的困難,也是必要解決的課題。本研究經由文獻回顧找出影響事件持續時間長短的主要因素,完成從不同機構取得的資料格式統一化與初步的篩選,其次使用資料融合技術進行資料的處理,讓多種來源的資料整合後具有說明事件發生的描述能力。接著分析造成事件的組成因子,包含時間、地點地理特性、參與事件的車輛類型與數量、事件的種類等因素,輔以車流量資料來確認影響事件持續時間的因子,再引用類神經網路模式來進行影響因子輸入與事件持續時間預測。對於輸出的結果採用MAPE指標來進行評估,俾以有效預測事件持續時間,案例數據顯示類神經網路模式架構對於事件延遲時間的存續有一定的預測能力。本研究除了針對事件延遲時間探討,也針對事件發生的時空背景來進行討論,試圖以多元的角度來進行事件延遲時間之預測工作。最後提供事件發生後如何進行延遲時間預測的應用程序,在事件發生時除了進行預測工作外並告知應該注意的相關事項。本研究準確性合理的結果可以作為事件處理單位客觀評估基礎,進而轉換為適當交通資訊給受事件影響後方駕駛人員,以期降低對交通之衝擊。
The forecast of accident duration is one of the critical steps in the accident management systems. Several approaches have been developed to predict the impact of accident duration. To provide useful traffic information requires the integration of several components such as software, hardware, and data collection systems.
Typically, these databases are maintained by different agencies for their own purposes, which are often incompatible. This research presents a model with the aim
of accident duration forecast based on historical datasets from Taiwan Freeway Systems.
This research utilizes the data fusion technology to combine several datasets, and to confirm the format of datasets compatible in the model. The duration forecast model is developed on Artificial Neural Networks model with significant factors affecting the range of duration. Results of the model yield reasonable predictions based on the criteria of mean absolute percentage error (MAPE) scale. Additionally, the test results of different situations in the duration forecast model are stable. The outputs of model may be used as the reference for traffic management, and to make rational diversion in the event of accident. Finally, such information will beneficially help drivers in mitigation traffic congestion during accident.
中文文獻
1.李穎,考慮事故延續時間動態更新預測之高速公路旅行時間預測模式建立與資料簡化方法比較,國立成功大學交通管理科學系,博士論文,民國96年。
2.林士傑,高速公路旅行時間預測模式之研究—類神經網路之應用,國立成功大學交通管理科學系,碩士論文,民國90年。
3.周文賢,多變量統計分析SAS/STAT使用方法,智勝文化事業有限公司,民國91年。
4.高速公路局,高速公路統計年報,民國97年。
5.陳順宇,迴歸分析,華泰書局,民國86年。
6.陳順宇,多變量分析,華泰書局,民國87年。
7.黃志偉,高速公路肇事處理時間預測之研究—應用類神經網路分析,國立中央大學土木工程學系,碩士論文,民國91年。
8.許正憲,資料融合技術應用於事故影響下高速公路旅行時間預測之研究,國立成功大學交通管理科學系,碩士論文,民國95年。
9.徐宏宇,以車流模擬模式探討高速公路旅行時間推估所需之客運GPS探針車數,國立成功大學電信管理研究所,碩士論文,民國95年。
10.葉怡成,應用類神經網路,儒林圖書公司,民國89年。
11.蘇木村、張孝德,機器學習類神經網路、模糊系統與基因演算法則,全華科技圖書股份有限公司,民國93年。
12.蔡宗憲、魏健宏,類神經網路於時間數列預測之發展趨勢回顧與展望
13.蔡宗憲、李治綱、魏健宏,短期列車旅運需求之類神經網路預測模式建構與評估,運輸計劃季刊,35卷4期,475-506頁,民國95年。
14.廖年慶,以基因演算法求解時間相依且隨機之最短路徑問題,國立成功大學工業管理科學系,碩士論文,民國94年。
15.魏健宏、林士傑、李穎,高速公路客運車輛旅行時間預測之實證評析,運輸計劃季刊,32卷4期,651-680頁,民國92年。
16.魏健宏、林士傑、李穎,高速公路事件延時特性分析-以中山高速公路北部路段為例,土木水利,30卷5期,85-88頁,民國92年。
17.魏健宏、陳奕志,類神經網路模式在國內交通運輸研究之成果評析,運輸計劃季刊,30卷2期,323-348頁,民國90年。
18.魏健宏、林鄉鎮,應用類神經網路與遺傳演算法建構小汽車跟車模式之研究,運輸計劃季刊,28卷3期,353-378頁,民國88年。
英文文獻
1.Chung, Y., “Development of an accident duration prediction model on the Korean Freeway Systems,” Accident Analysis & Prevention, Vol. 42, pp. 282-289, 2010.
2.Dailey, D. J., Harn, P. & Lin, P. J., “ITS Data Fusion,” Washington State Department of Transportation, 1996.
3.Dharia, A. & Adeli, H., “Neural network model for rapid forecasting of freeway link travel time,” Engineering applications of artificial intelligence, pp. 607-613, 2003.
4.Dougherty, M., “A review of neural networks applied to transport,” Transportation Research, Part C, Vol. 3, No.4, pp. 247-260, 1995.
5.Hall, D. L., “Mathematical techniques in Multisensor data fusion,” Artech House,Boston, 1992.
6.Ivan, J. N., “Real-time data fusion for arterial street incident detection using neural networks,” Transportation Research Record, No. 1497, pp. 27-35, 1995.
7.Ivan, J. N., “Neural network representations for arterial street incident detection data fusion,” Transportation Research, Part C, Vol. 5, No. 34, pp. 245-254, 1997.
8.Jia, Z., Balasuriya, A., & Challa, S., “Vision based data fusion for autonomous vehicles target tracking using interacting multiple dynamic models,” Computer Vision and Image Understanding, Vol.109, Issue 1, pp. 1-21, 2008.
9.Keay, K., & Simmonds, I., “Road accidents and rainfall in a large Australian city,” Accident Analysis & Prevention, Vol. 38, Issue 3, pp. 445-454, 2006.
10.Lee, Y. & Wei, C. H., “Data Fusion and Feature Composition Approach To Sequential Accident Duration Forecasting,” Journal of the East Asia Society for Transportation Studies, Vol. 7, pp. 2546-2560, 2007.
11.Leung, M.T., A.S. Chen & Daouk, H., “Forecasting exchange rate using general regression neural network,” Computers and Operations Research, pp.1093-1110, 2000.
12.Lo, H. K., Luo, X. W. & Siu, B. W. Y., “Degradable transport network: Travel time budget of travelers with heterogeneous risk aversion,” Transportation Research, Part B, Vol. 40, No.9, pp. 792-806, 2006.
13.Nelson, P. C., Application of neural networks to data fusion: a feasibility study,Transportation research board, 1996.
14.Sitzabee, W.E., Rasdorf, W., & Hummer, J.E., “Data Integration of Pavement Markings: A Case in Transportation Asset Management,” Journal of Computing in Civil Engineering, Vol.23, No.5, pp.288-298, 2009.
15.Tsai, T. H., Lee, C. K. & Wei, C. H., “Neural network based temporal feature models for short-term railway passenger demand forecasting,” Expert Systems with Applications, Vol. 36, No. 2P2, pp. 3728-3736, 2009.
16.Wei, C.H., & Lee, Y., “Sequential forecast of incident duration using artificial neural network models.”Accident Analysis & Prevention, Vol. 39, Issue 5, pp. 944-954, 2007
17.Wei, C. H. & Lee, Y., “Development of freeway travel time forecasting models by integrating different sources of traffic data,” IEEE Transactions on Vehicular Technology, Vol. 56, No. 6, November, pp. 3682-3694, 2007.
18.Zhoa, N., Yu, L., Geng, Y.B., & Chen X.M., “Support vector machine-based approach to data-layer multi-source ITS data fusion,” Journal of Transportation Systems Engineering and Information Technology, Vol.7, Issue 2, pp. 32-37, 2007.