Identification Methods of Critical Combination of Vulnerable Links in Transportation Networks

Abstract

This chapter presents two global optimization approaches for identifying the most critical combination of vulnerable links in a transportation network. The first approach is formulated with a bi-level framework. It can be applied to small-scale networks. The second proposed approach formulates this problem as a Mixed-integer Nonlinear Programming (MINLP). We revised the [MINLP] into a nicer form [RMINLP] to simplify the vulnerability problem into a discrete Network Design Problem (DNDP), so that it can be solved by many existing algorithms for DNDPs, and expand its scope of application. A numerical example of Sioux-Falls network is used to illustrate the two proposed approaches. The test results show that both approaches can provide a global optimization solution, while the second approach has a better property in computing efficiency.