Influence of Experience on Users’ Behaviour: A Day-to-Day Model for Route Choice Updating

AbstractThe performance of a transportation network is the consequence of users’ choices and the interaction of these choices. Several aspects influence users’ choices. For the analyst of the transportation system, being able to forecast users’ behaviour is necessary in order to obtain a simulation close to the reality, and in order to predict the impact of hypothetical policy measures. In private car transport networks, the route choice of drivers is one of the key choice dimensions to be considered. We suppose that three main elements influence route choice: i) the subjective perception of travel times, ii) experience directly obtained on the network by the user, and iii) information acquired through information systems like Advanced Traveler Information Systems (ATIS). In particular, in the present paper we delve into the influence of information on choices, by exploring the influence of the process by which the perception of travel times is updated day-by-day due to experience obtained on the network. In our mathematical model, the users’ awareness/knowledge of actual conditions on the network changes through an evolution process. Although this kind of problem has been previously studied, a novel feature of our approach is that as a consequence of actual experience or information, users may become aware of new alternatives: hence the ‘choice set’ perceived to be available may change day-by-day, not only the perception of travel times within a given choice set. In this paper, therefore, the time-evolution of both the choice (the alternative actually chosen) and the choice set (the alternative considered) is investigated. An illustrative specification of the model is proposed, and a series of simulation experiments conducted