A new approach to distributing urban trips

The traditional approach to urban transportation planning involves trip generation, distribution, modal split and assignment. For the most part these steps are logical and have a reasonable foundation in reality. The weakest of these on logical grounds is the distribution of trips using the gravity model. This approach gained acceptance in large part because total trips in an urban area tend to decrease with increasing distance. It is argued here that this is an artefact of the geographical distribution of trip productions and attractions. If this is true it has significant implications for sustainable transport, lands use planning and congestion management, i.e., we may be controlling the wrong variables. A new approach to trip distribution is proposed that assigns trips on the basis of opportunity sets (possible destinations) and trip purpose ranges (the maximum distance one is willing to or must travel for a given purpose).