Traffic congestion is one of the main problems in large cities. This problem stems from several factors such as: lack of parking space, people not using public traort and travel with personal cars. So far, several approaches have been presented to reduce traffic. Using the park and ride is an appropriate and efficient approach. Park and ride performance depends on it’s location on the urban network, and if these facilities are well-site then they have many benefits. Some of these benefits include: the reduction of fuel consumption, air pollution, travel time and congestion in downtown areas. In this research we want to propose a model to find the best loctions for park and ride establishment, so that this maximizes the network removed traffic. This model is formulated based on population areas, potential points for the establishment of the park and ride and several central business districts. In this study, the coefficient of travel satisfaction is considered and the time factor is employed to calculate network traffic. In other words, network traffic is equal to total time that is spent by all vehicles in the network. It’s proved that the model solving time that computed by full count method, follow exponential function. Therefore, we used metaheuristic algorithm for solving large size problems. Results that obtained, show the efficiency of this algorithm. Also the results show that the metaheuristic algorithm reachs the desired answer at the proper time. Finally, the model is used for locating park and ride in the city of Isfahan and it determined the proper locations for establishing park and rides and the optimum number of parks and rides have been identified. Keywords Park and ride, Facilitiy location, Flow capturing, Genetic algorithm