Reliability-based Dynamic Discrete Network Design with Stochastic Networks

Stochastic supply and fluctuating travel demand lead to stochastic travel times and travel costs for travelers. This paper will firstly focus on modeling of travelers' departure time/route choice behavior under stochastic capacities. By analytically proving the equivalency of the scheduling approach and the mean variance approach, a generalized travel cost function is derived to model travelers' departure time/route choice behavior under uncertainty. The proposed generalized travel cost function, which is more behaviorally sound and flexible, will be adopted to model a reliability-based long term user equilibrium with departure time choices. A reliability-based dynamic network design approach is proposed and formulated in which the numbers of lanes on all potential links are the design variables. A combined road network-oriented genetic algorithm and set evaluation algorithm is proposed to solve the dynamic network design problem. A new systematic approach is proposed to eliminate the infeasible, unrealistic and illogical lane designs in order to reduce the solution space and to save computation time. The proposed reliability-based dynamic network design approach is applied to a hypothetical network, and its solutions are compared to a corresponding static network design approach. It is concluded that the static network design approach may lead to poor designs. In general static traffic assignment underestimates the overall total network travel time and total network travel costs. The dynamic network design approach appears to result in a fairly good allocation of road capacity over space and makes the best utilization of the network capacity over time. A version of the Braess paradox appears in case of reliability-based cost functions in both static and dynamic networks.