Ehsan Ramezani Khansari - PhD student, Amirkabir University of Technology (Tehran Polytechnic),
Masoud Tabibi - Assistant Professor of Transportation Engineering, Civil Engineering Department, Engineering Faculty, Amirkabir University of Technology (Tehran Polytechnic)
Fereidoon Moghadas Nejad - Associate Professor of Transportation Engineering, Civil EngineeringDepartment, Engineering Faculty, Amirkabir University of Technology (Tehran Polytechnic)
Ehsan Amini - MSc of Transportation Engineering, Civil Engineering Department, Engineering Faculty, Amirkabir University of Technology (Tehran Polytechnic)

Estimation of the traffic flow s speed, in different week days and even different parts of a day, considered as one of the most important parameters for both traffic planners and road customers. The inaccuracy of traffic s data as well as the difference of driver s behavior, lead researchers to use the fuzzy for modeling the traffic events. In this study a mix of the fuzzy logic model and artificial neural network (adaptive neuro-fuzzy inference systems (ANFIS)) is used to model the speed of a traffic flow by using other traffic s data. At first, the fussy logic s roles are presented by an export person in traffic matter, and then these roles were rechecked and modified. The results show that the presented model can estimate the flow s speed with 10 Km/hour as statistical error

ANFIS, Modeling, Speed, Traffic flow