Policy-based stochastic dynamic traffic assignment models and algorithms

Abstract

Stochasticity is prevalent in transportation networks in general, and traffic networks in particular. We develop a policy-based stochastic dynamic traffic assignment (DTA) model and related solution algorithms. The DTA model works in a stochastic time-dependent network where link travel times are time-dependent random variables, Routing policies rather than paths are used as users' routing choices. A routing policy is a decision rule which specifies what node to take next out of current node based on current time and realized link travel times. We first give a conceptual framework for the DTA model. We then develop generic models for the routing policy generation problem, users' policy choice problem and dynamic network loading problem, which are the three major components of the overall DTA model. We then present a heuristic algorithm to solve the proposed policy-based DTA model. Using an example, we show that policy-based DTA models have solutions different, in expected travel times than the path-based models which are commonly used in the literature.