Chia-Ying Chien , 2006.06
Department of Transportation Technology and Management
National Chiao Tung University
Time dependent route guidance and travel estimation are important elements in a Advanced traveler information system（APTS）. Most previous researches focus on providing shortest paths; but neglect the occurrence probability. This shortage results in the extreme low probability of the provided route.
We propose three procedures to determine possible travel routes and their associated probabilities in a time-dependent network. The first and the second algorithm generate possible travel time range of the suggested path based on the algorithm proposed by Miller-Hooks and Mahmassani (1998). The third algorithm calculates the occurrence probabilities of a given route. The computation complexity of these three algorithms is also analyzed.
We use simulated data to evaluate performance of these three algorithms. The testing results shows that these algorithms are efficient and can provide useful information to travelers. |
