Urban public transit network optimization with flexible demand

Abstract

A new optimization approach is introduced for designing a public transit network. The main innovation of this paper is that the demand for public transit is endogenous. We maximize the total number of expected public transit passengers subject to a budget constraint. The demand depends on the established routes and their frequencies. A binary logit model is applied to simulate the behavior of the potential customers where the expected utility for public transit depends on in-vehicle time, waiting time, and access time. The expected waiting time depends on the frequency of the transit routes. The approach has been tested for the City of Dresden in Germany. The computational experiments show that the approach is applicable to real-world situations.