THESIS
2004
x, 68 leaves : ill. ; 30 cm
Abstract
Empirical studies have suggested that travel time variability plays an important role in travelers' route choice behavior. This thesis develops an approach to relate the travel time variability due to stochastic network link capacity variations and stochastic demand variations with travelers' risk aversive route choice behaviors. It is postulated that travelers acquire the variability of route travel times based on past experiences and factor such variability into their route choice consideration in the form of a travel time budget. This travel time budget varies with individuals and trip purposes and is related to the requirement on punctual arrivals. Moreover, all travelers want to minimize their travel time budgets. A multi-class mixed equilibrium mathematical program is formulated t...[
Read more ]
Empirical studies have suggested that travel time variability plays an important role in travelers' route choice behavior. This thesis develops an approach to relate the travel time variability due to stochastic network link capacity variations and stochastic demand variations with travelers' risk aversive route choice behaviors. It is postulated that travelers acquire the variability of route travel times based on past experiences and factor such variability into their route choice consideration in the form of a travel time budget. This travel time budget varies with individuals and trip purposes and is related to the requirement on punctual arrivals. Moreover, all travelers want to minimize their travel time budgets. A multi-class mixed equilibrium mathematical program is formulated to capture the route choice behaviors of travelers with heterogeneous risk aversions or requirements on punctual arrivals. Such an understanding has important implications on strengthening critical network links. Moreover the travel time budget model is extended to a logit-based route choice model which includes perception error of the travelers. This study then conducts numerical studies to illustrate the properties of the models.
Post a Comment