The Design of a Vehicle Traffic Flow Prediction Model for a Gauteng Freeway Based on an Ensemble of Multi-Layer Perceptron

Authors: Tebogo Emma Makaba, Barnabas Ndlovu Gatsheni

Abstract:

The cities of Johannesburg and Pretoria both located in the Gauteng province are separated by a distance of 58 km. The traffic queues on the Ben Schoeman freeway which connects these two cities can stretch for almost 1.5 km. Vehicle traffic congestion impacts negatively on the business and the commuter’s quality of life. The goal of this paper is to identify variables that influence the flow of traffic and to design a vehicle traffic prediction model, which will predict the traffic flow pattern in advance. The model will unable motorist to be able to make appropriate travel decisions ahead of time. The data used was collected by Mikro’s Traffic Monitoring (MTM). Multi-Layer perceptron (MLP) was used individually to construct the model and the MLP was also combined with Bagging ensemble method to training the data. The cross—validation method was used for evaluating the models. The results obtained from the techniques were compared using predictive and prediction costs. The cost was computed using combination of the loss matrix and the confusion matrix. The predicted models designed shows that the status of the traffic flow on the freeway can be predicted using the following parameters travel time, average speed, traffic volume and day of month. The implications of this work is that commuters will be able to spend less time travelling on the route and spend time with their families. The logistics industry will save more than twice what they are currently spending.

Keywords: Bagging ensemble methods, confusion matrix, multi-layer perceptron, vehicle traffic flow.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1338646

References:

[1] Gauteng Rail Link between Johannesburg, Pretoria and Johannesburg International Airport, Nov 01, 2014, pp. 1-12.
[2] A. Van Niekerk, Gauteng Freeway Improvement Project (GFIP) and Open Road Tolling. (PowerPoint Presentation). SANRAL (Mar 2013), pp.9 Accessed at: http://www.dbsa.org/en/prodserv/PACBP/MPIM% 20Documents/Gauteng%20Freeway%20Improvement%20Project%20and%20Open%20Road%20Tolling%20by%20Alex%20van%20Niekerk%20-%20SANRAL.pdf.
[3] Y.L Dahee Hong, Development of Traffic Accident Prediction Models. Proceedings of the Eastern Asia Society for Transportation Studies, 5, 2011, pp.2046-51.
[4] T. Thianniwet and S. Phosaard, Classification of Road Traffic Congestion Levels from GPS Data using a decision Tree Algorithm and Sliding Windows, IEEE Conference on intelligent trans. Sys., Proceedings of the 12th International , 2009 , pp. 978-988.
[5] J. He, Q. He, G. Swirszcz, Y. Kamarianakis, R. Lawrence, W. Shen, and L. Wynter,”Ensemble-based Method for Task 2:Predicting Traffic Jam”, Int. IEEE., Conf. on Data mining workshop New York 10598, 2010, pp.1363-1365.
[6] S. Mao, Y. Chen, and X. Xiao, Short-term Traffic Flow Prediction Based on GM (1, 1, Exp) model. International Conference on grey systems and intelligent services. (Sept. 2011), pp. 271-275.
[7] J. Yang , Travel Time Prediction Using the GPS Test Vehicle and Kalman Filtering Techniques., American control conference Portland, OR, USA, 8-10.Jun.2005, pp. 2128-2133.
[8] L. Chen and C.L. Phillip Chen, Ensemble Learning Approach for Freeway Short-Term Traffic Flow Prediction, IEEE, Department of Electrical and Computer engineering, The University of Texas at San Antonio, 2007, pp. 1-6.
[9] P. Pongpaibool, P. Tangamchit P. and Noodwong K., “Evaluation of traffic congestion using fuzzy techniques.” IEEE Trans., University of Technology Thouburi, Thailand, 2007, pp. 1-4.
[10] Jungme Park, Z.C.L.K.M.L.K.M.A.M., 2009. Intelligent Vehicle Power Control Based on Machine Learning of Optimal Control Parameters and Prediction of Road Type and Traffic Congestion. Vehicular Technology, IEEE Transactions on (Volume: 58, Issue: 9), 2009,pp 4741 -- 4756
[11] P. Le and W. Zuidema, 2014, Perceptron and Multi-layer Perceptron, International Conference on Measuring Technology and Mechatronics Automation, 2014, pp. 1-8.
[12] S. Haykin, neural network: A comprehensive foundation. New Jersey: Macmillan, 1994, pp.4-5.
[13] G. Cybenko, “Approximation by super positions of a sigmoidal function.” Math Contr. Signals Syst., vol. 2, 2003, pp. 303-314.
[14] M.C. Su, W.F. Jean, and H.T. Chang, " A Static Hand Gesture Recognition System Using a Composite Neural Network," in Fifth IEEE Int. Conf. on Fuzzy Systems, 1996, pp. 786- 792, New Orleans, U.S.A. (NSC85-2213-E-032-009).
[15] S. Beale and T. Jackson, Text to Phoneme Alignment and Mapping for Speech Technology: A Neural Networks Approach. In: Proceedings of the International Joint Conference on Neural Networks (IJCNN 2011), Bullinaria, 2011, 625-632. IEEE.
[16] H.H. Makwana, D.J. Shah, and P.P. Ghandhi, FPGA Implementation of Artificial Neural Network. International Journal of Emerging Technology and Advanced Engineering, January 2013, 3(1):672-679.
[17] J. Park, F. Diehl, M.J.F. Gales and M. Tomalin, 2009. Training and adapting MLP features for Arabic speech recognition. IEEE International Conference on Acoustics, Speech and Signal Processing, 19-24 April 2009:4461-4464.
[18] S. Russell and P. Norvig, 2003, Artificial Intelligence: A modern approach. 2nd ed, 2003.
[19] P. Tan, M. Steinbach and V. Kumar, Introduction to data mining: Pearson International Edition. 2006, pp 25-35.
[20] D.L. Hall, and J. Llinnas, “An introduction to multi-sensor data fusion,” in Proc. Of IEEE, vol. 85, no. 1, 1997, pp. 6-23.
[21] I. Syarif, E. Zaluska, A. Prugel-Bennett and G. Wills, 2012, Application of Bagging, Boosting and Stacking to Intrusion Detection, 2012.
[22] R.P. Sweety, Ensemble of classifier using artificial neural networks as base classifier, Int Journal of Comp Sci and Mob App, vol.1 issue. 4 Oct -2013, pp. 7-16
[23] C.M. Bishop, 1995.Neural Networks for Pattern Recognition. Oxford University Press (Month 1995).
[24] T. Chai and R.R. Draxler, 2014, Root Mean square error (RMSE) or Mean absolute error (MAE), “Arguments against avoiding RMSE in literature”, 2014, pp. 1-4.