Equitable Transit Network Design Under Uncertainty

TL;DR

This paper introduces a bilevel transit network design model that optimizes frequency settings to enhance efficiency and equity under supply uncertainty, revealing complex trade-offs and the impact of passenger risk attitudes.

Contribution

It develops a novel bilevel model incorporating supply uncertainty and analyzes the trade-offs between efficiency and equity in transit planning.

Findings

Pareto frontier may be non-convex

Simultaneous improvements in efficiency and equity are possible

Increasing frequency can worsen equity measures

Abstract

This paper proposes a bilevel transit network design problem considering supply side uncertainty. The upper level problem determines frequency settings to simultaneously maximize the efficiency and equity measures, which are defined by the reduction in the total effective travel cost and the minimum reduction in the effective travel cost of all OD pairs, respectively. The lower level problem is the reliability based transit assignment problem that captures the effects of supply-side uncertainty on passengers route choice behavior.Numerical studies demonstrate that 1) the Pareto frontier may not be convex; 2) it is possible to improve the efficiency and equity objectives simultaneously; 3) increasing the frequency could worsen the equity measure; 4) passengers risk attitude affects the rate of substitution between the two objectives.