In this paper a method for predicting the proportions of trips falling within peak periods is described and some applications of the method within the Netherlands National Model. The National Model is used to investigate long-term transportation policy for the strategic road and railway networks. Many proposed policies (e.g. road pricing) vary with the time of day, as does the congestion on the network. To investigate the effectiveness of the policies in changing the time-of-day distribution of travel, as well as to incorporate a realistic representation of the different conditions at different times, models have been developed that predict the proportion of travel that will occur in each peak period. The paper indicates how models of time-of-day choice, operating at the level of origin-destination movements and exhibiting different time and cost sensitivities by purpose, can be integrated with the existing models of mode and destination choice. Brief mention is made of the stated preference methods that were used to derive these models and the models themselves are presented. Finally, the paper describes the integration of the resulting mode-destination-time-of-day structure with the assignment procedures, which measure congestion and derive the link-specific road pricing. Solution methods for the equilibration procedures are also described with reference to specific policy investigations.
Abstract