YAPAY SİNİR AĞLARI ve DESTEK VEKTÖR MAKİNELERİ YÖNTEMLERİ ile BÖLGESEL TRAFİK YOĞUNLUK TAHMİNİ

Büyük şehirlerde kilometre başına düşen insan yoğunluğu arttıkça trafik sıkışıklığı artmakta ve yolcuların daha fazla sürelerini trafikte harcamaktadırlar. Trafik sıkışıklığı nedeni ile harcanan ekstra zaman ve yakıt hem kullanıcılar hem de ülkeler için büyük bir gider kalemidir. Büyükşehirlerde yaşayan vatandaşların trafik yoğunluğunun zaman bazlı değişimini tahmin etmek ve buna göre planlama yapmaları bir zorunluluk haline dönüşmüştür. Trafik sıkışıklıkları genelde tüm şehirde aynı anda gerçekleşmez. Bölgesel olarak yaşanan trafik sıkışıklıkları diğer yolları da etkilemesi ile yaygınlaşır. Bu çalışma da yapay sinir ağları (YSA) kullanılarak önerilen yöntem ile geçmiş trafik verileri kullanarak bölgesel yoğunluklar tahmin edilmeye çalışılacaktır. Çalışma birçok benzer çalışmadan farklı olarak hava durumu gibi çevresel etkenleri de alarak tahmin modellemesinin başarısını arttırılmıştır. İstanbul Büyük Şehir Belediyesi Açık Veri Portalından toplanan 75 farklı noktaya ait 150.000 veri kullanarak önerilen model test edilmiş ve yaklaşık %90 başarı ile bölgesel trafik yoğunluğu tespit edilebilmiştir.

Anahtar Kelimeler:

Trafik Tahmini, YSA, SVM, Ortalama Hız, Büyük Veri.

PREDICTION of REGIONAL TRAFFIC INTENSITY with ARTIFICIAL NEURAL NETWORKS and SUPPORT VECTOR MACHINES

As the density of people per kilometer increases in big cities, traffic congestion increases and passengers spend more time in traffic. The extra time and fuel spent due to traffic congestion is a big expense item for both users and countries. It has become a necessity to predict the time-based change in the traffic density of citizens living in metropolises and to plan accordingly. Traffic jams don't usually happen in the whole city at once. Regional traffic jams become widespread as they affect other roads. In this study, it will be tried to predict regional congestions by using historical traffic data with the proposed method using artificial neural networks (ANN). The study increases the success of forecasting modeling by taking environmental factors such as weather conditions apart from many equivalent studies. Using 150,000 data from 75 different points collected from the Istanbul Metropolitan Municipality Open Data Portal, the proposed model was tested and the regional traffic density could be determined with 90% success.

Keywords:

Traffic Prediction, ANN, SVM, Average Speed, Big Data,

PDF

___

[1] SputnikNews TR, “İstanbullular trafikte ne kadar vakit kaybediyor? - Sputnik Türkiye.” https://tr.sputniknews.com/analiz/201712201031487131-istanbullular-trafikte-ne-kadar-vakitkaybediyor/ (accessed Jun. 17, 2021).
[2] INRIX, “Scorecard - INRIX,” 2020. Accessed: Mar. 18, 2021. [Online]. Available: https://inrix.com/scorecard/.
[3] E. Romanova, “Increase in Population Density and Aggravation of Social and Psychological Problems in Areas with High-Rise Construction,” in E3S Web of Conferences, Mar. 2018, vol. 33, p. 03061, doi: 10.1051/e3sconf/20183303061.
[4] M. E. Hallenbeck, J. M. Ishimaru, and J. Nee, “MEASUREMENT OF RECURRING VERSUS NON-RECURRING CONGESTION,” Washington (State). Dept. of Transportation, Oct. 2003. Accessed: Mar. 18, 2021. [Online]. Available: https://rosap.ntl.bts.gov/view/dot/38804.
[5] “FHWA Operations - Reducing Recurring Congestion.” https://ops.fhwa.dot.gov/program_areas/reduce-recur-cong.htm (accessed Mar. 18, 2021).
[6] N. J. Mazzenga Graduate, R. Assistant, and M. J. Demetsky, “Investigation of Solutions to Recurring Congestion on Freeways Virginia Transportation Research Council,” Virginia Transportation Research Council, 2009. Accessed: Mar. 18, 2021. [Online]. Available: http://www.virginiadot.org/vtrc/main/online_reports/pdf/09-r10.pdf.
[7] “Welcome to ROSA P | Welcome.” https://rosap.ntl.bts.gov/ (accessed Mar. 19, 2021).
[8] F. Sun, A. Dubey, and J. White, “DxNAT - Deep neural networks for explaining non-recurring traffic congestion,” in Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017, Jul. 2017, vol. 2018-January, pp. 2141–2150, doi: 10.1109/BigData.2017.8258162.
[9] F. Sun, A. Dubey, and J. White, “DxNAT - Deep neural networks for explaining non-recurring traffic congestion,” in Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017, Jul. 2017, vol. 2018-Janua, pp. 2141–2150, doi: 10.1109/BigData.2017.8258162.
[10] S. G. Farrag, F. Outay, A. U.-H. Yasar, and M. Y. El-Hansali, “Evaluating Active Traffic Management (ATM) Strategies under Non-Recurring Congestion: Simulation-Based with Benefit Cost Analysis Case Study,” Sustainability, vol. 12, no. 15, p. 6027, Jul. 2020, doi: 10.3390/su12156027.
[11] T. M. Brennan, R. A. Gurriell, A. J. Bechtel, and M. M. Venigalla, “Visualizing and Evaluating Interdependent Regional Traffic Congestion and System Resiliency, a Case Study Using Big Data from Probe Vehicles,” J. Big Data Anal. Transp., vol. 1, no. 1, pp. 25–36, Jun. 2019, doi: 10.1007/s42421-019-00002-y.
[12] C.-L. Lan, R. Venkatanarayana, and M. D. Fontaine, “Development of a Methodology for Determining Statewide Recurring and Nonrecurring Freeway Congestion: Virginia Case Study,” Transp. Res. Rec. J. Transp. Res. Board, vol. 2673, no. 6, pp. 566–578, Jun. 2019, doi: 10.1177/0361198119850471.
[13] E. Kidando, R. Moses, T. Sando, and E. E. Ozguven, “Evaluating Recurring Traffic Congestion using Change Point Regression and Random Variation Markov Structured Model,” Transp. Res. Rec. J. Transp. Res. Board, vol. 2672, no. 20, pp. 63–74, Dec. 2018, doi: 10.1177/0361198118787987.
[14] H. Nguyen, W. Liu, and F. Chen, “Discovering Congestion Propagation Patterns in SpatioTemporal Traffic Data,” IEEE Trans. Big Data, vol. 3, no. 2, pp. 169–180, Jul. 2016, doi: 10.1109/tbdata.2016.2587669.
[15] S. S. Anjum et al., “Modeling Traffic Congestion Based on Air Quality for Greener Environment: An Empirical Study,” IEEE Access, vol. 7, pp. 1–24, 2019, doi: 10.1109/ACCESS.2019.2914672.
[16] S. Amini, N. Motamedidehkordi, E. Papapanagiotou, and F. Busch, “Estimation of traversal speed on multi-lane urban arterial under non-recurring congestion,” in 5th IEEE International Conference on Models and Technologies for Intelligent Transportation Systems, MT-ITS 2017 - Proceedings, Aug. 2017, pp. 514–519, doi: 10.1109/MTITS.2017.8005726.
[17] Y. Wu, H. Tan, L. Qin, B. Ran, and Z. Jiang, “A hybrid deep learning based traffic flow prediction method and its understanding,” Transp. Res. Part C Emerg. Technol., 2018, doi: 10.1016/j.trc.2018.03.001.
[18] R. Fu, Z. Zhang, and L. Li, “Using LSTM and GRU neural network methods for traffic flow prediction,” 2017, doi: 10.1109/YAC.2016.7804912.
[19] H. Yi, J. Heejin, and S. Bae, “Deep Neural Networks for traffic flow prediction,” 2017, doi: 10.1109/BIGCOMP.2017.7881687.
[20] B. Sharma, S. Kumar, P. Tiwari, P. Yadav, and M. I. Nezhurina, “ANN based short-term traffic flow forecasting in undivided two lane highway,” J. Big Data, vol. 5, no. 1, p. 48, Dec. 2018, doi: 10.1186/s40537-018-0157-0.
[21] B. Gültekin Çetiner, M. Sari, and O. Borat, “A neural network based traffic-flow prediction model,” Math. Comput. Appl., vol. 15, no. 2, pp. 269–278, 2010, doi: 10.3390/mca15020269.
[22] F. Canitez, P. Alpkokin, and S. T. Kiremitci, “Sustainable urban mobility in Istanbul: Challenges and prospects,” Case Stud. Transp. Policy, vol. 8, no. 4, pp. 1148–1157, Dec. 2020, doi: 10.1016/j.cstp.2020.07.005.
[23] T. T. Yaman, H. B. Sezer, and E. Sezer, “Modeling Urban Traffic by Means of Traffic Density Data: Istanbul Case,” in Advances in Intelligent Systems and Computing, Jul. 2021, vol. 1197 AISC, pp. 867–874, doi: 10.1007/978-3-030-51156-2_100.
[24] F. Canitez and M. Deveci, “An Integration Model for Car Sharing and Public Transport : Case of Istanbul,” Transist Istanbul Transp. Congr. Exhib., no. April, pp. 1–10, 2017, Accessed: Mar. 20, 2021. [Online]. Available: https://www.researchgate.net/publication/324530842.
[25] “Istanbul Metropolitan Municipality Air Quality Station Information Web Service.” https://data.ibb.gov.tr/tr/dataset/hava-kalitesi-istasyon-bilgileri-web-servisi (accessed Feb. 26, 2021).
[26] D. F. Specht, “A General Regression Neural Network,” IEEE Trans. Neural Networks, vol. 2, no. 6, pp. 568–576, 1991, doi: 10.1109/72.97934.
[27] A. S. Ahmad et al., “A review on applications of ANN and SVM for building electrical energy consumption forecasting,” Renewable and Sustainable Energy Reviews, vol. 33. Elsevier Ltd, pp. 102–109, 2014, doi: 10.1016/j.rser.2014.01.069.
[28] R. Grosse, “Lecture 5: Multilayer Perceptrons.”
[29] F. Wahid, R. Ghazali, A. S. Shah, and M. Fayaz, “Prediction of Energy Consumption in the Buildings Using Multi-Layer Perceptron and Random Forest,” Int. J. Adv. Sci. Technol., vol. 101, pp. 13–22, Apr. 2017, doi: 10.14257/IJAST.2017.101.02.
[30] “Electrical load forecasting using support vector machines | IEEE Conference Publication | IEEE Xplore.” .
[31] O. Nelles, “Classical Polynomial Approaches,” Nonlinear Syst. Identif., pp. 893–901, 2020, doi: 10.1007/978-3-030-47439-3_20.
[32] I. Ebtehaj, H. Bonakdari, A. H. Zaji, H. Azimi, and F. Khoshbin, “GMDH-type neural network approach for modeling the discharge coefficient of rectangular sharp-crested side weirs,” Eng. Sci. Technol. an Int. J., vol. 18, no. 4, pp. 746–757, Dec. 2015, doi: 10.1016/J.JESTCH.2015.04.012.