Importance sampling for activity path choice

摘要: We propose a model for the choice of an activity pattern. Models participation patterns allow to assess impact demand management strategies on and destination choices. In particular, we focus set generation using recent developments in route modeling. Spatial choices deal with large sets. develop framework based path choice. The activity-episode sequence is modeled as network defining type, duration time day. dimensionality managed through importance sampling Metropolis-Hastings algorithm. Our can be used forecast at urban scale also pedestrian facilities, such transport hubs or mass gathering. Validation approach performed synthetic data case study WiFi traces campus presented.