This thesis investigates the role of travel information on traveller's route choice behaviour based on revealed preference (RP) data collected in a congested network. A modelling framework denominated Recursive Logit Dynamic extends the Recursive Logit to a dynamic setting by taking the variability of traffic conditions into account. This is particularly relevant when dealing with real time travel information. In such modelling framework the route choice problem is formulated as sequential link choice decisions that can be consistently estimated based on path observations. (Author/publisher)
Samenvatting