ARIMA和深度神經網路的混合模型預測eTag偵測率

簡易檢索 / 詳目顯示

回結果列表

研究生：	邱正穎 Chiou, Cheng-Ying
論文名稱：	ARIMA和深度神經網路的混合模型預測eTag偵測率 A hybrid ARIMA and Deep Neural Network model to forecast eTag detection rate
指導教授：	蘇朝墩 Su, Chao-Ton
口試委員:	陳穆臻 Chen, Mu-Chen 蕭宇翔 Hsiao, Yu-Hsiang 薛友仁 Shiue, Yeou-Ren
學位類別：	碩士 Master
系所名稱：	工學院 - 工業工程與工程管理學系 Department of Industrial Engineering and Engineering Management
論文出版年：	2020
畢業學年度：	108
語文別：	英文
論文頁數：	66
中文關鍵詞：	電子收費系統、時間序列預測、自回歸移動平均模型法、深度神經網絡法
外文關鍵詞：	Electronic toll collection system, Time series forecasting, ARIMA, Deep neural network
相關次數：	點閱：4 下載：0
分享至:	分享至facebook 分享至twitter

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

近年來，電子收費系統逐漸取代了傳統的人工收費。台灣的ETC主要由感應器和RFID eTag組成，而隨著使用時間的增加，車燈型eTag的偵測率會降低。因此，本研究針對eTag使用時間進行偵測率預測。
本研究使用台灣ETC的數據進行時間序列預測，使用自回歸移動平均模型法，深度神經網絡法，和將上述兩種方法結合起來的混合法進行預測。最後得出結論混合法比其他方法具有更好的效果。根據最後的預測結果，本研究建議ETC公司使用兩年後應該召回客戶以更換新的eTag。
通過此研究中的分析，可以減少未檢測到的車輛數量。而該公司也可以藉此減少其人力成本或時間。

In recent years, the electronic toll collection system has gradually replaced traditional manual toll collection. The ETC in Taiwan is mainly composed of a reader and an RFID eTag. With the increasing of used time, the detection rate of headlight type eTag will become lower. Therefore, we develop a forecasting model based on eTag used time and detection rate.
This study used the data of Taiwan ETC for time series forecasting. We applied ARIMA method, deep neural network method, and hybrid method which combined above two methods to conduct prediction. We conclude that hybrid method performs better than else two and recommend that ETC company should recall the customers to replace a new eTag after using two years.
Through our analysis in this research, the number of undetected vehicles can be decreased. As a result, the company can reduce its operation costs or time.

   Introduction.........................................7
1    Research Background and Motivation...................7
2    Research Purposes....................................8
3    Research Architecture................................8
   Literature Review...................................10
1    Electronic Toll Collection System...................10
2    Time Series Forecasting.............................11
2.1    Direct Multi-step Forecast Strategy.................11
2.2    Recursive Multi-step Forecast Strategy..............12
2.3    Direct-Recursive Hybrid Strategy....................12
2.4    Multiple Output Strategy............................13
3    Applications of Time Series Forecasting.............13
4    ARIMA method........................................15
5    Deep Neural Network.................................16
6    Hybrid method.......................................17
   Research methodology................................19
1    ARIMA method........................................19
2    Deep Neural Network method..........................24
3    Proposed Hybrid method..............................26
   Case study..........................................28
1    The problem.........................................28
2    Data collection.....................................29
3    ARIMA method........................................32
4    DNN method..........................................39
5    Proposed Hybrid method..............................49
6    Comparison .........................................52
7    Result and Recommendation...........................54
   Conclusion .........................................56
1    Conclusion..........................................56
2    Future study........................................56
References..................................................58
Appendix....................................................61

                                

[1] Aladag, C. H., Egrioglu, E., & Kadilar, C. (2009). Forecasting nonlinear time series with a hybrid methodology. Applied Mathematics Letters, 22(9), 1467-1470.
[2] Areekul, P., Senjyu, T., Toyama, H., & Yona, A. (2009). Notice of violation of ieee publication principles: A hybrid arima and neural network model for short-term price forecasting in deregulated market. IEEE Transactions on Power Systems, 25(1), 524-530.
[3] Bontempi, G., Taieb, S. B., & Le Borgne, Y. A. (2012, July). Machine learning strategies for time series forecasting. In European business intelligence summer school (pp. 62-77). Springer, Berlin, Heidelberg.
[4] Chatfield, C. (2000). Time-series forecasting. CRC press.
[5] Chen, P. C., Gadepalli, K., MacDonald, R., Liu, Y., Kadowaki, S., Nagpal, K., Kohlberger, T., Dean, J., Corrado, G. S., Hipp, J. D., Mermel, C. H., & Stumpe, M. C. (2019). An augmented reality microscope with real-time artificial intelligence integration for cancer diagnosis. Nature medicine, 25(9), 1453–1457.
[6] De Gooijer, J. G., & Hyndman, R. J. (2006). 25 years of time series forecasting. International journal of forecasting, 22(3), 443-473.
[7] FETC. https://www.fetc.net.tw/.
[8] Frank, R. J., Davey, N., & Hunt, S. P. (2001). Time series prediction and neural networks. Journal of intelligent and robotic systems, 31(1-3), 91-103.
[9] Freeway Bureau, MOTC. https://www.freeway.gov.tw/.
[10] Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: principles and practice. OTexts.
[11] Jason Brownlee. (2017). 4 strategies for multi-step time series forecasting. https://machinelearningmastery.com/multi-step-time-series-forecasting/
[12] Karsoliya, S. (2012). Approximating number of hidden layer neurons in multiple hidden layer BPNN architecture. International Journal of Engineering Trends and Technology, 3(6), 714-717.
[13] Kim, K. J. (2003). Financial time series forecasting using support vector machines. Neurocomputing, 55(1-2), 307-319.
[14] Kumar, M., & Thenmozhi, M. (2012). Stock Index Return Forecasting and Trading Strategy Using Hybrid ARIMA-Neural Network Model. International Journal of Financial Management, 2(1).
[15] Masum, S., Liu, Y., & Chiverton, J. (2018, June). Multi-step time series forecasting of electric load using machine learning models. In International Conference on Artificial Intelligence and Soft Computing (pp. 148-159). Springer, Cham.
[16] Moghadam, R. A., & Keshmirpour, M. (2011, November). Hybrid ARIMA and neural network model for measurement estimation in energy-efficient wireless sensor networks. In International Conference on Informatics Engineering and Information Science (pp. 35-48). Springer, Berlin, Heidelberg.
[17] Ouyang, S.Y. (2019). Machine Learning Techniques for Vehicle Detection based on ETC data. National Tsing Hua University Industrial Engineering and Engineering Management, Master Thesis.
[18] Siami-Namini, S., Tavakoli, N., & Namin, A. S. (2018, December). A comparison of ARIMA and LSTM in forecasting time series. In 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA) (pp. 1394-1401). IEEE.
[19] Su, C. T. (2013). Quality engineering: off-line methods and applications. CRC press.
[20] Taieb, S. B., Bontempi, G., Atiya, A. F., & Sorjamaa, A. (2012). A review and comparison of strategies for multi-step ahead time series forecasting based on the NN5 forecasting competition. Expert systems with applications, 39(8), 7067-7083.
[21] Taieb, S. B., & Hyndman, R. J. (2012). Recursive and direct multi-step forecasting: the best of both worlds (Vol. 19). Department of Econometrics and Business Statistics, Monash Univ..
[22] Tang, Z., & Fishwick, P. A. (1993). Feedforward neural nets as models for time series forecasting. ORSA journal on computing, 5(4), 374-385.
[23] Tiao, G. C., & Tsay, R. S. (1994). Some advances in non‐linear and adaptive modelling in time‐series. Journal of forecasting, 13(2), 109-131.
[24] Tsai, R.S. (2019). Using Machine Learning Approaches for Improving the Performance of RFID-based Vehicle Detection Rate. National Tsing Hua University Industrial Engineering and Engineering Management, Master Thesis.
[25] Wongsathan, R., & Seedadan, I. (2016). A hybrid ARIMA and Neural Networks model for PM-10 pollution estimation: The case of Chiang Mai city moat area. Procedia Computer Science, 86, 273-276.
[26] Zhang, G.P. (2003). Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing, 50, 159-175.

簡易檢索 / 詳目顯示

相關論文