Optimal frequency, route and fare structure for transit networks with elastic demand

Abstract

Public transit is a very profitable and competitive industry in Hong Kong and the passengers’ route choice and mode choice behavior will be significantly influenced by the level of services provided, thus it is crucial for the transit operators to determine their frequency, route and fare structure in order to maximize their own profit. In this paper a bi-level model is developed to optimize the route structure, frequency setting and fare structure simultaneously. The upper level problem is to maximize the operator’s profit while the lower level problem is an approach-based SUEED transit assignment model. An artificial bee colony (ABC) algorithm is proposed to solve determine the route structure and a sensitivity analysis-based algorithm is proposed to find out the optimal frequency and fare settings. Numerical examples with discussions will be presented to investigate the behavior of the model and to examine the effectiveness of the proposed algorithm. Comparison on three different fare structures, i.e., flat fare, distance based fare and sectional fare is made.