Selecting and Scheduling Link and Intersection Improvements in Urban Networks

Deciding which projects, alternatives or investments to implement is a complex and important problem not only in transportation engineering, but in management, operations research and economics. Projects are interrelated if their benefits or costs depend on which other projects are implemented. Furthermore, in the network development problem analyzed here, the timing of projects also affects the benefits and costs of other projects. This paper presents a method for optimizing the selection and scheduling of interrelated improvements in road networks that explicitly considers intersections. The Frank Wolfe algorithm, which is modified here to consider intersections, is used for evaluating network improvements as well as for traffic assignment. Intersections are modelled with pseudo-links whose delays are estimated with Akcelik's generalized model. The objective is to minimize the present value of total costs (including user time) by determining which projects should be selected and when they should be completed. A genetic algorithm is used for optimizing the sequence and schedule of projects.