An Activity Based Demand Model for Large Scale Simulations.

Abstract This paper presents the ongoing development of SCAPER, a random utility based travel demand model that consistently incorporates time decisions. The paper focusses on how SCAPER can be used for large scale simulations, and more specifically: 1. How computational speed of SCAPER is improved using sampling of locations, and how it influences the simulation results. 2. Interfacing SCAPER to MATSim simulation framework, and estimating the SCAPER model with travel times and travel costs produced by the simulation of Stockholm demand (simulated) over Stockholm network using MATSim. 3. Preliminary results from 1 and 2.