Accessibility-based Network Design Model Considering Induced Travel Demands

Abstract

The paramount goal of transportation planning is to provide sufficient accessibility for citizens to access various urban services in their daily life. However, few network design models have been developed to design network improvement schemes with the objective of maximizing total system accessibility. To fill this gap, we propose a new accessibility-based discrete network design model to maximize the accessibility in determining the optimal network design scheme. The proposed model is formulated as a bi-level optimization problem. The upper level is formulated as an accessibility maximization problem subject to the construction budget constraint. The lower level is formulated as an equivalent elastic-demand user equilibrium model. To handle the proposed model, an effective solution algorithm is developed by integrating genetic algorithm with Frank–Wolfe algorithm. Numerical examples using Sioux Falls network are carried out to demonstrate the merits of the proposed model by comparing to the conventional mobility-based network design models.