This paper deals with an analysis and comparison of individual routes, surveyed and modelled, of car trips in an urban road network. The objective is to assess the predictive quality of the so-called shortest time route choice model for car drivers. One of the findings is that the minimum-time route choice model gives a fairly good prediction of actual routes: nearly 50% of the routes is predicted correctly. Especially the part of the routes on the more important roads is predicted very well.
Abstract