Revelation of preferences in trip distribution models

Abstract

In this paper we will see how commuter preferences can be revealed from observations of trip distributions. We will explain how to find unique representations of preference structures leading to an observed trip distribution, and will also present a numerical method that can be used to infer preferences from systems of considerable size. The theory is applied to a real world network, and we show how our framework can be used to reveal detailed information about the spatial structure of this network. We also use this new framework to suggest a new approach to evaluate the impact of road pricing.