A novel doubly stochastic transit assignment formulation: An application to singapore bus network

Abstract

This paper develops a Variational Inequality (VI) formulation for the proposed doubly stochastic transit assignment problem considering the risk-taking behaviour of passengers in a congested transit network. In-vehicle travel time, waiting time, capacity, and in-vehicle congestion are considered as stochastic variables simultaneously and their influence on the overall cost function is modelled. A new cost function is also proposed to capture the stochastic capacity of a route section. Passengers' perception errors on transit network conditions are considered in the VI formulation and the reliability-based stochastic user equilibrium condition is defined. Numerical results are reported for Singapore bus network. Sensitivity analyses of parameters involved are also discussed in detail.