A Trip Generation Method for Time-Dependent Large-Scale Simulations of Transport and Land-Use

This paper presents an attempt to integrate dynamic traffic models and location choice models in a more consistent way than the four-step planning scheme. Two temporal horizons are considered. On the long run, individuals select the location of their primary activity based on local land-use data and on travel costs. Mode choice is omitted. On the short run, car travelers select the departure time and route on the journey to their the primary activity. On the short run, the travel demand is disaggregated at the individual level. On the long run, users characteristics are aggregated. A methodology is developed that can be used to bootstrap multi-agents simulations of land-use and transport in an efficient way. The emphasis is put on the provision of an operational model. Therefore, the paper focuses on the technical details of the implementation of this methodology and on its applications to large-scale systems. Results are presented on the Zurich area.