In this study, the road traffic congestion of Dehradun city is evaluated from traffic flow information using fuzzy techniques. Three different approaches namely Sugeno, Mamdani models which are manually tuned techniques, and an Adaptive Neuo-Fuzzy Inference System (ANFIS) which an automated model decides the ranges and parameters of the membership functions using grid partition technique, based on fuzzy logic. The systems are designed to human’s feelings on inputs and output levels. There are three levels of each input namely high, medium and low for input density, fast, medium and slow for input speed, and five levels of output namely free flow, slow moving, mild congestion, heavy congestion and serious jam for the road traffic congestion estimation. The results, obtained by fuzzy based techniques show that the manually tuned Sugeno type technique achieves 72.05% accuracy, Mamdani type technique achieves 83.82% accuracy, and Adaptive Neuro-Fuzzy Inference System technique achieves 88.23% accuracy. ANFIS technique appears better than the manually tuned fuzzy technique, and also the manually tuned fuzzy technique gives good accuracy which leads that the fuzzy inference system can capture the human perception better through manual adjustment of input/output membership functions.
P. Posawang, S. Phosaard, W. Polnigongit and W. Pattara-Atikom, “Perception-Based Road Traffic Congestion Classification Using Neural Networks,” Proceedings of the World Congress on Engineering, London, 1-3 July 2009.
A. Kablan, “Adaptive Neuro-Fuzzy Inference System for Financial Trading Using Intraday Seasonality Observa- tion Model,” World Academy of Science, Engineering and Technology, Vol. 58, No. , 2009, pp. 479-488.
F. Porikli and X. Li, “Traffic Congestion Estimation Us- ing HMM Models without Vehicle Tracking,” IEEE In- telligent Vehicles Symposium, Parma, 14-17 June 2004, pp. 188-193. doi:10.1109/IVS.2004.1336379
B. Krause and C. von Altrock, “Intelligent Highway by Fuzzy Logic: Congestion Detection and Traffic Control on Multi-Lane Roads with Variable Road Signs,” 5th In- ternational Conference on Fuzzy Systems, New Orleans, 8-11 September 1996, pp. 1832-1837.
A. S. Alfa, “Understanding Urban Traffic Congestion dur- ing Peak Periods,” Proceedings of International Confer- ence on Road and Road Transport Problems (ICORT-88), Roorkee, 12-15 December 1988, pp. 518-527.
S. M. Seyedhoseini, “Application of Adaptive Neuro-Fuz- zy Inference System in Measurement of Supply Chain Agility: Real Case Study of a Manufacturing Company,” African Journal of Business Management, Vol. 4, No. 1, 2010, pp. 83-96.
T. Takagi and M. Sugeno, “Derivation of Fuzzy Control Rules from Human Operator’s Control Actions,” Pro- ceedings of IFAC Symposium on Fuzzy Information, Knowledge Representation and Decision Analysis, Mar- seilles, 19-21 July 1983, pp. 55-60.
N. Patchanee, P. Tangamchit and P. Pongpaibool, “Road Traffic Estimation from a GPS-Equipped Car Using Fuz- zy Logic,” Proceeding of 29th Electrical Engineering Con- ference, Chonburi, 9-10 November 2006, pp. 1081-1084.