Abstract
Present traffic network model require that all possible origins and destinations of trips be represented as if they were taking place to and from a small set of points or centroids. This misrepresents the access cost from trip-ends to the networks, especially the cost from home to train stations. In order to alleviate this spatial aggregation problem, we suggest a new network model which regarding connecter cost as random valuables. We applied this model to the railway networks in the Tokyo Metropolitan Area, and confirmed its applicability