(1) 由以上結果發現，短距離旅行時間預測之誤差約為 6.5%而長距離旅行 時間預測之誤差約為 7.1%，因此本方法在兩個及三個收費站間旅行時間之 預測績效沒有太大的差異。
(2) 由於本研究目前沒有演算法直接根據交通資訊判斷是否該時段發生事 故，因此在資料過濾時，即把一些速度過慢、旅行時間過長之情形判斷為 離群值而進行過濾，因此未來如果加入事件資料庫，可以增加判斷事件發 生的機制，而建立在事件發生時之旅行時間資料庫，將有助於在事故發生 時進行旅行時間預測。
(3) 本研究目前對於參數的設定如權重的設定、門檻值的設定及距離量度的 給定，由於目前沒有一篇論文是完全利用 k-NN 方法進行旅行時間預測，並 且目前參數值均以比較方法進行搜索，像是時間門檻的設定，本研究根據 平均旅行時間大部分落於(42~64 分間)，故本研究以半小時做為時間的門檻，
找尋在每半小時下，最接近的歷史資料。希望未來可以有進一步的研究，
探討這些參數要如何決定。
(4) 目前本研究僅以歷史資料進行過濾及插補，然而若應用於實際案例中，
如果偵測器有離群值的產生，或是資料缺漏，就無法得到準確的資料。
(5) 未來可以再多取一些樣本，驗證是否本模式有足夠的解釋意義。
參 考 文 獻
[1] Wachs, M.,“Relationships between drivers’ attitudes toward alternate routes and driver and route characteristics”,Highway Research Record , vol.197, pp.70-87, 1967.
[2] van Berkum, E., and P. van der Mede, “The Impact of Traffic Information: Dynamics in Route and Departure Time Choice, Delft University Press, Delft, the Netherlands”, PhD. thesis of the Transport and Planning Department, Faculty of Civil Engineering and Geosciences, Delft Unive rsity of Technology, 1993.
[3] 運輸安全組，“北臺灣科技走廊智慧型系統規劃與開發”，交通部運 輸研究所， 2008 年 6 月。
[4] Kwon, J., B. Coifman, and P. Bickel,“Day-to-Day Travel-Time Trends and Travel-Time Prediction from Loop-Detector Data” , Transportation Research Record: Journal of the Transportation Research Board, Vol.1717,pp.120-129,2000.
[5] Sen, A., P. Thakuriah, and X.Q. Zhu, “Frequency of Probe Reports and Variance of Travel Time Estimates”, Journal of Transportation Engineering, Vol. 123, No. 4, pp. 290-297, 1997.
[6] Rice, J. and E. van Zwet, “A Simple and Effective Method for Predicting Travel Times on Freeways”, IEEE Transactions on Intelligent Transportation Systems, Vol. 5(3), pp. 200-207, 2004.
[7] Yang,J.S. and D. Billings , “Application of The ARIMA Models to Urban Roadway Travel Time Prediction”,SMC IEEE
International Conference, Vol.3, pp.2529-2534, 2006.
[8] Vapnik, V., “Predicting Time Series with Support Vector Machine”, Springer LNCS 1327, pp.999-1004.
[9] Wu, C.H., J.M. Ho, and D.T. Lee, “Travel time prediction with support vector regression”, IEEE Transactions on Intelligent Transportation Systems, Vol.5, NO. 4, pp. 276-281, 2004.
[10] Yoshikazu, O., K. Toshihiro, and S. Shigehito,
“Online-Learning Type of Traveling Time Prediction Model in Expressway”, Intelligent Transportation System, ITSC IEEE Conference, pp. 350-355, 1998.
[11]李季森，「應用探測車法預測高速公路旅行時間」，中央大學土木工程 研究所，碩士論文，2001。
[12]張修榕，「高速公路旅行時間之研究」，中央大學土木工程研究所，碩 士論文，2001。
[13]黃裕文，「高速公路施工路段旅行時間預測之研究」，中央大學土木工 程研究所，碩士論文，2003。
[14]溫志元，「高速公路進口匝道匯流路段旅行時間研究」，中央大學土木 工程研究所，碩士論文，2002。
[15]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」，成 功大學交通管理科學研究所，碩士論文，2001。
[16]吳佳峰，「有 GPS 資訊提供下之車輛旅行時間預估模式之研究」，交通 大學運輸工程與管理系，碩士論文，2001。
[17]李穎，「國道客運班車旅行時間預測模式之研究」，成功大學交通管理 科學研究所，碩士論文，2002。
[18]張慶麟，「應用自動車輛辨識預測高速公路路段旅行時間」，中央大學 土木工程研究所，碩士論文，2002。
[19]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，第 32 卷，第 4 期，651~679 頁，2003。
[20] Chang, G.L., T. Junchaya, and A.J. Santiago, “A real-Time Network Traffic Simulation Model For ATMS Applications: Part I – Simulation Methodologies”, Journal of Intelligent
Transportation Systems, pp. 227-241, 1994.
[21] Johnston, C.M., and A.T. Chronopoulos, “The parallelization of a Highway Traffic Flow Simulation”, Frontiers of Massively Parallel Computation, pp.192-199, 1999.
[22] Kachani, S., and G. Perakis,“Second-Order Fluid Dynamics Models for Travel Travel Times in Dynamic Transportation Networks”, Proceedings on IEEE Intelligent Transportation Systems Conference, pp. 251-256, 2001
[23] Kiesling, T., and J. Luthi, “Towards Time-Parallel Road Traffic Simulation,” Principles of Advanced and Distributed Simulation, pp. 7-15, 2005.
[24] Benedetti, J., “On the Nonparametric Estimation of Regression Functions”, Journal of the Royal Statistical Society, Series B, Vol. 39, pp. 248-253, 1977.
[25] Stone, C., “Consistent Nonparametric Regression”, The Annals of Statistics, Vol. 5, pp. 595-645, 1977.
[26] Tukey, J., “Exploratory Data Analysis”, Reading, MA: Addison Wesley, 1977.
[27] Altman, N.S., “An Introduction to Kernel and Nearest Neighbor Nonparametric Regression”, The American Statistician, Vol. 46, NO. 3, pp. 175-185, 1992.
[28] Davis, G.A., and N.L. Nihan, “Nonparametric Regression and Short-Term Freeway Traffic Forecasting”, Journal of
Transportation Engineering, Vol. 117, NO. 2, pp. 178-188, 1991.
[29] Smith, B., and M. Demetsky, “Traffic Flow Forecasting:
Comparison of Modeling Approaches”, Journal of Transportation Engineering, Vol.123, NO.4, pp. 261–266, 1997.
[30] Smith, B., B. Williams, and K. Oswald, “Comparison of Parametric and Non-parametric Models for Traffic Flow
Forecasting”, Transportation Research Part C, Vol. 10(4), pp.
303–321, 2002.
[31] Clark, S., “Traffic Prediction using Multivariate Nonparametric Regression”, Journal of Transportation Engineering, Vol. 129(2), pp. 161-168, 2003.
[32] Robinson, S., and J. Polak, “Modeling Urban Link Travel Time with Inductive Loop Detector Data by Using the k-NN Method”, Journal of the Transportation Research Board, Vol. 1935, pp. 47-56, 2005.
[33] Chang, G.L., N. Zou, and J. Wang, “Development and Field Evaluation of a Real-Time Travel Time Prediction System”, Maryland Department of Transportation State Highway
Administration, Final Report, 2006.
[34] Fukunaga,K., and L. Hostetler ,“Optimization of
k-Nearest-Neighbor Density Estimates ,”IEEE Transactions on Information Theory, Vol.19,pp.320-326, 1973.
[35] 交通部統計處，“交通統計月報”，交通部，2008。
附錄一 各偵測器交通資訊之信賴區間
0 120 240 400 520 640 800 920 1040 1200 1320 1440 1600 1720 1840 2000 2120 2240
每
0 125 250 415 540 705 830 955 1120 1245 1410 1535 1700 1825 1950 2115 2240
每
-50
0 120 240 400 520 640 800 920 1040 1200 1320 1440 1600 1720 1840 2000 2120 2240
每
0 125 250 415 540 705 830 955 1120 1245 1410 1535 1700 1825 1950 2115 2240
每
-50
0 120 240 400 520 640 800 920 1040 1200 1320 1440 1600 1720 1840 2000 2120 2240
每
0 125 250 415 540 705 830 955 1120 1245 1410 1535 1700 1825 1950 2115 2240
每
-50
0 125 250 415 540 705 830 955 1120 1245 1410 1535 1700 1825 1950 2115 2240
每
0 125 250 415 540 705 830 955 1120 1245 1410 1535 1700 1825 1950 2115 2240
每
-50
0 120 240 400 520 640 800 920 1040 1200 1320 1440 1600 1720 1840 2000 2120 2240
每
0 125 250 415 540 705 830 955 1120 1245 1410 1535 1700 1825 1950 2115 2240
每
-50
0 120 240 400 520 640 800 920 1040 1200 1320 1440 1600 1720 1840 2000 2120 2240
每
0 125 250 415 540 705 830 955 1120 1245 1410 1535 1700 1825 1950 2115 2240
每
-50
0 120 240 400 520 640 800 920 1040 1200 1320 1440 1600 1720 1840 2000 2120 2240
每
0 125 250 415 540 705 830 955 1120 1245 1410 1535 1700 1825 1950 2115 2240
每
-50
0 120 240 400 520 640 800 920 1040 1200 1320 1440 1600 1720 1840 2000 2120 2240
每
0 125 250 415 540 705 830 955 1120 1245 1410 1535 1700 1825 1950 2115 2240
每
-100
0 120 240 400 520 640 800 920 1040 1200 1320 1440 1600 1720 1840 2000 2120 2240
每
0 125 250 415 540 705 830 955 1120 1245 1410 1535 1700 1825 1950 2115 2240
每
附錄二 VD 與 ETC 推估旅行時間比較圖
12 13 14 15 16 17 18
00:00 01:10 02:20 03:30 04:40 05:50 07:00 08:10 09:20 10:30 11:40 12:50 14:00 15:10 16:20 17:30 18:40 19:50 21:00 22:10 23:20
5/12(二)
ETC VD
12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 17
00:00 01:10 02:20 03:30 04:40 05:50 07:00 08:10 09:20 10:30 11:40 12:50 14:00 15:10 16:20 17:30 18:40 19:50 21:00 22:10 23:20
5/13(三)
ETC VD
12 13 14 15 16 17 18
00:00 01:10 02:20 03:30 04:40 05:50 07:00 08:10 09:20 10:30 11:40 12:50 14:00 15:10 16:20 17:30 18:40 19:50 21:00 22:10 23:20
5/18(一)
ETC VD
12 13 14 15 16 17 18 19 20 21 22
00:00 01:10 02:20 03:30 04:40 05:50 07:00 08:10 09:20 10:30 11:40 12:50 14:00 15:10 16:20 17:30 18:40 19:50 21:00 22:10 23:20
5/19(二)
ETC VD
12 13 14 15 16 17 18 19 20 21 22
00 :10 20 30 40 50 00 10 20 30 40 :50 00 10 20 30 40 50 00 10 20
5/20(三)
ETC VD
12 13 14 15 16 17 18 19 20 21 22
00:00 01:10 02:20 03:30 04:40 05:50 07:00 08:10 09:20 10:30 11:40 12:50 14:00 15:10 16:20 17:30 18:40 19:50 21:00 22:10 23:20
5/22(五)
ETC VD
12 14 16 18 20 22 24
00:00 01:10 02:20 03:30 04:40 05:50 07:00 08:10 09:20 10:30 11:40 12:50 14:00 15:10 16:20 17:30 18:40 19:50 21:00 22:10 23:20
5/24(日)
ETC VD
12 14 16 18 20 22 24
00:00 01:10 02:20 03:30 04:40 05:50 07:00 08:10 09:20 10:30 11:40 12:50 14:00 15:10 16:20 17:30 18:40 19:50 21:00 22:10 23:20
5/25(一)
ETC VD
附錄三 旅行時間預測結果整理
k-NN 模式旅行時間預測整理(970303(一))
時間 預測值 (分)
實際值 (分)
誤差值 (分)
誤差率
(%) 變異數 05:00 59.68293 57.19167 2.491261 4.36% 40.64368 06:00 63.25098 57.46481 5.786163 10.07% 76.91641 07:00 65.30463 62.635 2.669632 4.26% 104.4071 08:00 74.24098 81.79902 -7.55804 9.24% 79.51625 09:00 63.80902 59.44722 4.3618 7.34% 87.85285 10:00 61.92293 61.61204 0.310889 0.50% 51.35536 12:00 61.46049 57.55833 3.902154 6.78% 49.65348 13:00 60.40829 59.55 0.85829 1.44% 17.21197 14:00 59.38634 60.65655 -1.27021 2.09% 9.552464 15:00 60.34634 59.20833 1.138009 1.92% 9.814049 16:00 61.96195 61.44095 0.521006 0.85% 34.5413 17:00 62.22366 66.0358 -3.81214 5.77% 35.97134 18:00 62.32585 60.68653 1.639323 2.70% 20.35702 19:00 63.43122 63.07972 0.351497 0.56% 59.46309 20:00 61.81756 57.32143 4.496135 7.84% 65.11891 21:00 59.77585 60.35 -0.57414 0.95% 58.04229 22:00 59.50829 52.88667 6.621624 12.52% 57.93485 23:00 53.24927 54.87917 -1.6299 2.97% 89.74966
k-NN 模式旅行時間預測整理(970304(二))
時間 預測值 (分)
實際值 (分)
誤差值 (分)
誤差率
(%) 變異數 05:00 59.47756 56.29166 3.185897 5.66% 31.12386 06:00 62.16195 57.80833 4.353617 7.53% 46.89345 07:00 62.51463 75.03999 -12.5254 16.69% 70.23805 08:00 63.62829 74.25445 -10.6262 14.31% 98.8784 09:00 63.23927 58.13461 5.104655 8.78% 87.59885 10:00 62.83171 57.63333 5.198376 9.02% 83.20334 11:00 61.46024 70.71439 -9.25415 13.09% 49.92535 12:00 60.10707 59.41459 0.692488 1.17% 10.85384 13:00 60.40829 67.81333 -7.40504 10.92% 17.21197 14:00 59.38634 56.44295 2.94339 5.21% 9.552464 15:00 59.4022 72.63712 -13.2349 18.22% 10.47545 16:00 60.82512 61.01743 -0.1923 0.32% 19.10081 17:00 61.56561 67.64052 -6.07491 8.98% 20.81106 18:00 62.27854 63.04219 -0.76365 1.21% 20.13344 19:00 63.63585 61.59391 2.041944 3.32% 28.1084 20:00 62.88195 57.92578 4.956174 8.56% 89.60308 21:00 59.50829 56.49889 3.009407 5.33% 57.93485
k-NN 模式旅行時間預測整理(970305(三))
時間 預測值 (分)
實際值 (分)
誤差值 (分)
誤差率
(%) 變異數 05:00 60.14854 61.24444 -1.09591 1.79% 43.48507 06:00 61.2722 68.26111 -6.98891 10.24% 46.68024 07:00 64.2778 64.85834 -0.58053 0.90% 107.9457 08:00 65.52463 61.6375 3.887137 6.31% 217.803 09:00 63.23927 59.8375 3.401766 5.69% 87.59885 10:00 63.19073 61.10568 2.08505 3.41% 81.27873 11:00 61.50927 56.53944 4.969824 8.79% 49.50127 12:00 60.39366 63.03 -2.63634 4.18% 12.1592 13:00 60.40829 58.77917 1.629126 2.77% 17.21197 14:00 59.30146 57.37889 1.922576 3.35% 9.339123 15:00 60.2 59.61593 0.584071 0.98% 8.427365 16:00 61.85902 59.31088 2.548145 4.30% 33.82436 17:00 62.35341 70.79547 -8.44206 11.92% 57.8961 18:00 62.32585 59.18652 3.139331 5.30% 20.35702 19:00 63.43122 59.78662 3.644599 6.10% 59.46309 20:00 62.29561 63.2753 -0.97969 1.55% 63.93373 21:00 60.33049 55.30129 5.029203 9.09% 65.26557
k-NN 模式旅行時間預測整理(970308(六))
時間 預測值 (分)
實際值 (分)
誤差值 (分)
誤差率
(%) 變異數 05:00 59.67732 64.51667 -4.83935 7.50% 34.23878 06:00 62.92585 55.43333 7.49252 13.52% 74.24427 07:00 63.48805 71.93 -8.44195 11.74% 99.41214 08:00 63.62829 60.55227 3.07602 5.08% 98.8784 09:00 63.23927 75.35152 -12.1122 16.07% 87.59885 10:00 63.19073 64.09461 -0.90388 1.41% 81.27873 11:00 61.46049 60.94392 0.516572 0.85% 49.65348 12:00 60.3022 65.91349 -5.61129 8.51% 13.25599 13:00 60.8378 63.00206 -2.16425 3.44% 34.39317 14:00 59.30146 71.44 -12.1385 16.99% 9.339123 15:00 60.2 60.92981 -0.72981 1.20% 8.427365 16:00 61.96195 62.53651 -0.57456 0.92% 34.5413 17:00 62.34976 63.66306 -1.3133 2.06% 35.96177 18:00 62.32585 69.64307 -7.31721 10.51% 20.35702 19:00 63.98463 60.05606 3.928573 6.54% 57.8174 20:00 61.81756 57.39223 4.425334 7.71% 65.11891 21:00 59.78268 67.9803 -8.19762 12.06% 56.70156 22:00 59.50829 54.95548 4.552814 8.28% 57.93485 23:00 53.29341 57.56012 -4.26671 7.41% 90.06584
k-NN 模式旅行時間預測整理(970316(日))
時間 預測值 (分)
實際值 (分)
誤差值 (分)
誤差率
(%) 變異數 05:00 59.51843 56.68889 2.829542 4.99% 54.73985 06:00 59.58098 55.23333 4.347646 7.87% 55.50018 07:00 60.14941 56.73333 3.416078 6.02% 81.77313 08:00 60.20216 64.02055 -3.8184 5.96% 78.47968 09:00 63.03804 62.34528 0.692759 1.11% 87.59885 10:00 62.4998 58.9837 3.5161 5.96% 81.94361 11:00 61.01176 64.24514 -3.23338 5.03% 49.65348 12:00 60.18314 59.40857 0.774568 1.30% 10.85384 13:00 58.63373 57.95139 0.682336 1.18% 9.255779 14:00 59.56922 57.04548 2.523737 4.42% 9.339123 15:00 60.86608 65.39675 -4.53067 6.93% 8.427365 16:00 61.83902 62.48769 -0.64867 1.04% 34.5413 17:00 62.57333 67.6551 -5.08176 7.51% 35.96177 18:00 62.44118 66.9667 -4.52552 6.76% 20.35702 19:00 63.55843 69.38439 -5.82596 8.40% 57.8174 20:00 61.35157 61.23906 0.112513 0.18% 65.11891 21:00 59.45588 56.77148 2.684405 4.73% 58.04229 22:00 59.38078 56.23139 3.149392 5.60% 6.054254 23:00 54.17157 54.74088 -0.56931 1.04% 90.06584
簡 歷
姓名：蔡繼光
籍貫：臺灣省基隆市
生日：民國 72 年 2 月 16 日
學歷：民國九十八年七月國立交通大學運輸科技與管理學系碩士班畢業 民國九十四年七月國立交通大學運輸科技與管理學系學士班畢業 電郵信箱：bright7722@gmail.com