Simulating endogenous dynamics of activity-travel behavior: results of numerical simulations

Modelling the dynamics of activity-travel behavior constitutes the next challenge on the international research agenda. Long-term dynamics relate to changes in opportunities or constraints, such as residential and job choice, while short-term decisions relate to day-to-day experiences. Due to a bottom-up process, a constant series of negative experiences may trigger people to conduct long-term changes. The aim of the current study is to model the bottom-up process of influence into an agent-based simulation system of dynamic activity-travel behavior. This paper will discuss the conceptual framework underlying the model, as well as the results of numerical simulations. Specifically, the dynamic evolution of the choice-set that each agent generates for every activity type, at every time step will be simulated. Furthermore, the variation of the aspiration values that the agents hold for every activity type will be indicated. Finally, examples of short-term and long-term dynamics in agents' activity-travel behavior will be illustrated.