Bike network design problem with a path-size logit-based equilibrium constraint: Formulation, global optimization, and matheuristic

Abstract

This study focuses on the optimal network design problem of bike paths, which are on or adjacent to roadways but are physically separated from motorized traffic within the existing urban network. The problem seeks to maximize the total route utilities of cyclists and capture their actual route choice behavior using a path-size logit model. A mixed-integer nonlinear nonconvex model is developed for the problem and is reformulated and linearized into a mixed-integer linear program. The program is solved with a global optimization method and a matheuristic. Results are provided to illustrate the performance of these methods and the model properties.