Abstract
This paper addresses a maximal transit reserve trip generation which quantitatively reflects interaction between land-use and public transit system. It firstly proposes a variational inequality and a diagonalization method for the combined transit trip distribution/logit-based stochastic user equilibrium transit assignment with elastic frequencies of transit lines. Secondly, this paper develops a generalized bilevel programming model for the maximal transit trip generation, which consists of two level problems — upper and lower level problems. The upper level problem aims to maximize the reserve transit trip for each origin subject to the transit equity constraints, which is formulated as a multi-objective maximization problem. The lower level problem is the proposed variational inequality for the combined trip distribution/logit-based stochastic user equilibrium transit assignment. A genetic algorithm embedded with the diagonalization method is designed for solving the generalized bilevel programming model. Finally, a numerical example is carried out.
