عنوان مقاله [English]
The transportation planning process for urban areas is typically based on partition of the area into traffic zones. In transportation planning and network analysis, it is customary to use centroid nodes to represent traffic analysis zones (TAZ). In order to load trips of each zone on the roadway network, each centroid is connected to the roadway network by at least one connector. Globally there are few studies examining the impact of connectors on traffic assignment, but all of them emphasize the importance of this subject. In this paper, the effect of selecting connector nodes on traffic assignment results is studied and a new method is proposed to select connector nodes. To implement the concepts presented in a real sample, City of Mashad, Iran, was selected as the case study. The proposed method in this study emphasizes the proximity of connector node to centroid and other criteria. This proposed method is compared with two methods from the literature. In the first method the only criterion is that the connector node must be close to the centroid. The second method introduces a measure of weight for all nodes and then selects connector nodes. To evaluate the three methods, results of traffic assignment were compared to observations of traffic volume on 87 selected links. Evaluation of results shows an improvement in the goodness of fit index of the proposed method of 3.78 and 22.2 percent compared to the first and second methods, indicating more precision in replicating real world observations.