Planning is an integral part of many behavioural aspects related to transportation: residential relocation, activity and travel scheduling, route choice, etc. People make plans and then select actions to execute those plans. The plans are inherently dynamic. They evolve due to situational constraints and contextual factors, experience, inertia, or changing preferences. As a result, the chosen actions might be different from those initially planned. In this paper, we present the methodology to model the dynamics of choices using a two-layer decision hierarchy (choice of a plan and choice of action conditional on the plan) and its dynamics. This framework, based on Hidden Markov Model principles, assumes that the plan at every time period depends on the plan at the previous time period and the actions taken in the previous time periods as well as other variables including the characteristics of the decision maker. The dynamics in the observed actions are explained by the dynamics in the underlying latent (unobserved) plans. The methodology is demonstrated by modelling the dynamics associated with the driving decisions as the drivers enter a freeway. The model is estimated using disaggregate trajectory data and validated in a microscopic traffic simulator. - Planning is an integral part of many behavioural aspects related to transportation: residential relocation, activity and travel scheduling, route choice, etc. People make plans and then select actions to execute those plans. The plans are inherently dynamic. They evolve due to situational constraints and contextual factors, experience, inertia, or changing preferences. As a result, the chosen actions might be different from those initially planned. In this paper, we present the methodology to model the dynamics of choices using a two-layer decision hierarchy (choice of a plan and choice of action conditional on the plan) and its dynamics. This framework, based on Hidden Markov Model principles, assumes that the plan at every time period depends on the plan at the previous time period and the actions taken in the previous time periods as well as other variables including the characteristics of the decision maker. The dynamics in the observed actions are explained by the dynamics in the underlying latent (unobserved) plans. The methodology is demonstrated by modelling the dynamics associated with the driving decisions as the drivers enter a freeway. The model is estimated using disaggregate trajectory data and validated in a microscopic traffic simulator.
