(3.236.118.225) 您好!臺灣時間:2021/05/16 14:33
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

: 
twitterline
研究生:劉俐宣
研究生(外文):Li-Hsuan Liu
論文名稱:高速公路服務區之需求預測
論文名稱(外文):Forecasting Demand for Highway Service Areas
指導教授:蔡玫亭蔡玫亭引用關係
口試委員:王建富李際偉
口試日期:2016-06-21
學位類別:碩士
校院名稱:國立中興大學
系所名稱:企業管理學系所
學門:商業及管理學門
學類:企業管理學類
論文種類:學術論文
論文出版年:2016
畢業學年度:104
語文別:中文
論文頁數:59
中文關鍵詞:需求預測簡單貝氏分類法大數據
外文關鍵詞:Demand PredictionNaïve Bayes ClassifierBig Data
相關次數:
  • 被引用被引用:1
  • 點閱點閱:133
  • 評分評分:
  • 下載下載:17
  • 收藏至我的研究室書目清單書目收藏:3
隨著高速公路走向休閒娛樂化,每個服務區除了是基本休息場所之外,更是一個各具特色之景點。截至目前,服務區已提供越來越多樣的服務以滿足不同顧客,且年營業額和來客數都逐年成長甚至創新高,已是不可忽視之新商機。因此,本研究欲建構高速公路服務區需求預測模型來幫助服務區業者提高營業效率和營收。
本研究使用國道高速公路局資料庫之ETC車輛交易紀錄作為資料來源,利用簡單貝氏分類法為基礎來發展需求預測模型,以預測高速公路服務區之需求量。
本研究以清水服務區為對象,採用2015年7月之ETC車輛交易紀錄作為訓練集來訓練模型,再採用8月之資料作為驗證集來驗證模型的預測能力。本研究將車輛依旅次起點所在地,由北至南分段成多個區域,利用離散實證分配的方式計算條件機率,結果發現在雲嘉南地區和高屏地區的預測能力最好。而其他區域透過重新再次分段後,發現苗栗地區和彰化南投地區亦得到較好的預測效果。而若以服務區業者的角度,只衡量進入服務區的數量準確率,則新竹地區進入服務區之筆數準確率有達82.87%。由此可見,本研究之需求預測模型除了可利用大數據發揮預測的即時性,提供服務區業者作為營運方面像是食材存放控管和行銷活動的參考依據,亦可由數值分析中看出本研究之模型在大部分區域皆可有較高的預測能力。此外,本研究模型所使用的簡單貝氏分類法在使用上較為簡單、容易上手,尤其若要應用在大數據上面,此方法假設特徵變數之間相互獨立,可降低許多運算成本。

In recent years, highway service areas have been transformed into popular leisure spots. In addition to providing drivers places for rest, highway service areas offer diversified services to attract different consumers. Up until now, the total revenue and the average daily customer flow of these service areas have been growing drastically every year, and a business opportunity is definitely present. Therefore, in this study, a demand forecasting model is established for highway service areas in order to help operators improve business efficiency and increase revenue.
The data of this study was retrieved from Taiwan Area National Freeway Bureau, which was used to establish the demand forecasting model based on Naïve Bayes classifier.
The Cingshuei service area was chosen as our research object. The ETC traffic data of July 2005 was used to train the forecasting model, and the data of August 2015 was used to confirm the predictive validity. In this study, trip going from the norther part to southern part of Taiwan were segmented into several sub-data sets starting from drivers’ trip origin. Discrete empirical distribution was then adopted to compute conditional probability in this study, and the result indicated that demand forecasting model has good predictive validity in the area of Yunlin, Chiayi, Tainan, Kaohsiung and Pingtung. As other areas are also segmented into a few sub-data sets, we found that demand forecasting model also has good predictive validity in the area of Miaoli, Changhua and Nantou. However, compare to the number of vehicles by estimated value and actual value only, the accuracy of demand forecasting model in Hsinchu is 82.87%. According to the outcome, demand forecasting model shows good predictive validity in most of the area and can be applied to real-time prediction in order to help operators develop strategies. Moreover, Naïve Bayes classifier dependence assumptions between features to avoid the curse of dimensionality can reduce computing costs especially for big data.

第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 5
1.3 研究方法 5
1.4 研究流程 6
第二章 文獻探討 8
2.1 高速公路相關預測 8
2.2 分類法 14
2.2.1 階層式分類法 15
2.2.2 簡單貝氏分類法 18
2.3 貝氏分類在預測之應用 20
2.3.1 貝氏分類在運輸產業之預測 20
2.3.2 貝氏分類在大數據之預測 22
第三章 國道服務區車次預測模型 24
3.1 需求預測模型建立 24
3.2 預測模型之評估指標 28
第四章 數值分析 31
4.1 資料說明 31
4.2 清水服務區之需求量預測 33
4.2.1 資料處理過程 34
4.2.2 不分區之服務區需求預測 36
4.2.3 分區之服務區需求預測 39
4.2.4 針對特定區域作再分段之服務區需求預測 45
第五章 結論與建議 49
5.1 結論 49
5.2 研究限制與未來建議 50
參考文獻 51
附錄 …………………………………………………………………54

一、中文部分
交通部高速公路局(民104年)。各類車輛通行原始檔(運輸協會)。交通部高速公路局,民105年2月14日,取自:http://goo.gl/nvYyPA
馬岳琳、謝明玲(民101年7月10日)。特色台式服務 引爆旅途商機。天下雜誌501期,取自:http://www.cw.com.tw/article/article.action?id=5041857
郭昌儒(民104年7月24日)。探勘大數據─高速公路Big Data之分析應用。交通部研討會,取自:http://bigdata.iot.gov.tw/index.php?option=com_attachments&task=download&id=7
許慧美、吳權原、陳柏璋、金哲生(民103年2月4日)。國道湧收假車潮 清水服務區賺最多。三立新聞網,取自:http://www.setn.com/News.aspx?NewsID=12424
蔡偉祺(民104年4月6日)。設備好 景觀美 服務區人氣旺。中時電子報,取自:http://www.chinatimes.com/newspapers/20150406000278-260102
鄭瑋奇(民104年7月2日)。關廟服務區 明全新開幕。台灣新生報,取自:http://61.222.185.194/?FID=6&CID=278435
蘇瑋璇(民104年7月2日)。來去國道玩一天 服務區變特色景點。聯合報,取自:http://goo.gl/GpxKlV
二、西文部分
DUDA, R. O., HART, P. E. & STORK, D. G. 2012. Pattern Classification, John Wiley & Sons.
GORDON, A. D. 1999. Classification, Chapman & Hall, CRC, Boca Raton, FL.
MITCHELL, T. M. 1997. Machine Learning, McGraw-Hill, Inc.
THEODORIDIS, S. & KOUTROUMBAS, K. 2001. Pattern Recognition and Neural Networks. Machine Learning and Its Applications. Springer.
THEODORIDIS, S., PIKRAKIS, A., KOUTROUMBAS, K. & CAVOURAS, D. 2010. Introduction to Pattern Recognition: A Matlab Approach: A Matlab Approach, Academic Press.
ABBASI, A., ALBRECHT, C., VANCE, A. & HANSEN, J. 2012. MetaFraud: A Meta-Learning Framework for Detecting Financial Fraud. MIS Quarterly, 36, 1293-1327.
BARCELO, J., MONTERO, L., BULLEJOS, M., SERCH, O. & CARMONA, C. 2013. A Kalman Filter Approach for Exploiting Bluetooth Traffic Data When Estimating Time-Dependent OD Matrices. Journal of Intelligent Transportation Systems, 17, 123-141.
BREIMAN, L. 2001. Random Forests. Machine Learning, 45, 5-32.
BURGES, C. J. 1998. A Tutorial on Support Vector Machines for Pattern Recognition. Data Mining and Knowledge Discovery, 2, 121-167.
JUNQUE DE FORTUNY, E., MARTENS, D. & PROVOST, F. 2013. Predictive Modeling with Big Data: Is Bigger Really Better? Big Data, 1, 215-226.
HUANG, Y. J., POWERS, R. & MONTELIONE, G. T. 2005. Protein NMR Recall, Precision, and F-measure Scores (RPF scores): Structure Quality Assessment Measures Based on Information Retrieval Statistics. Journal of the American Chemical Society, 127, 1665-1674.
KARYPIS, G., HAN, E.-H. & KUMAR, V. 1999. Chameleon: Hierarchical Clustering Using Dynamic Modeling. Computer, 32, 68-75.
KOHAVI, R. Scaling Up the Accuracy of Naive-Bayes Classifiers: A Decision-Tree Hybrid. KDD, 1996. Citeseer, 202-207.
KUSAKABE, T. & ASAKURA, Y. 2014. Behavioural Data Mining of Transit Smart Card Data: A Data Fusion Approach. Transportation Research Part C: Emerging Technologies, 46, 179-191.
LEE, S., HEYDECKER, B., KIM, Y. H. & SHON, E. Y. 2011. Dynamic OD Estimation Using Three Phase Traffic Flow Theory. Journal of Advanced Transportation, 45, 143-158.
LIN, N., ZONG, C., TOMIZUKA, M., SONG, P., ZHANG, Z. & LI, G. 2014. An Overview on Study of Identification of Driver Behavior Characteristics for Automotive Control. Mathematical Problems in Engineering, 2014.
LIN, P.-W. & CHANG, G.-L. 2007. A Generalized Model and Solution Algorithm for Estimation of the Dynamic Freeway Origin–Destination Matrix. Transportation Research Part B: Methodological, 41, 554-572.
LOCKAMY III, A. & MCCORMACK, K. 2012. Modeling Supplier Risks Using Bayesian Networks. Industrial Management & Data Systems, 112, 313-333.
QUINLAN, J. R. 1986. Induction of Decision Trees. Machine learning, 1, 81-106.
QUINLAN, J. R. 1996. Improved Use of Continuous Attributes in C4. 5. Journal of Artificial Intelligence Research, 77-90.
WU, X., KUMAR, V., QUINLAN, J. R., GHOSH, J., YANG, Q., MOTODA, H., MCLACHLAN, G. J., NG, A., LIU, B. & PHILIP, S. Y. 2008. Top 10 Algorithms in Data Mining. Knowledge and Information Systems, 14, 1-37.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top