Semi-dynamic traffic assignment model with mode and route choices under stochastic travel times

Transportation network conditions vary significantly during the course of a day. In many urban areas, public transit and (private) automobiles constitute the actual forms of transportation that use such networks. Public transportation by rail is more reliable than by automobiles or buses; therefore, ordinary static and deterministic traffic assignment models with combined mode and route choices may not be suitable to assess a transportation network that includes public railways. Moreover, within-day dynamics and reliability need to be incorporated in such a model. In this paper, we use a semi-dynamic traffic assignment model that considers within-day dynamics by improving the static traffic assignment model. In addition, stochastic travel times are incorporated into the model. Thus, we propose a semi-dynamic traffic assignment model with mode choice between public transit and automobiles, route choice with stochastic travel times, and an accompanying computing algorithm. This model enables us to assess within-day dynamics of transportation networks and travel time reliability of public railways. Copyright © 2012 John Wiley & Sons, Ltd.