Hybrid transit system for transit metropolis: Is “spatiotemporal” hybrid structure better?

Abstract

This paper proposes a Hybrid Transit System for Transit Metropolis (HTS-TM) over a grid of streets which considers a cross-shaped urban development area, enabling a “spatiotemporal” hybrid network structure. The system divides an urban area into the central area and periphery which adopts different transit line spacing (i.e. spatially hybrid) and headway (i.e. temporally hybrid), while a grid transit network is employed citywide. The proportion of the central area and periphery can be optimized, and the flexible network structure can accommodate a centripetal demand pattern that has a large proportion of trips destinating for the central business district. We build a strategic planning model for HTS-TM to provide evidence for futuristic urban expansion from the perspective of transit development, or to serve as indicators for transit metropolis assessment. Specifically, we decompose the spatially heterogeneous demand into two layers and adopt a parsimonious continuum approach, computing the user and agency costs based on the layers. By striking a balance between the user and agency costs, the strategic planning problem is formulated as a mixed integer program with only a few decision variables. Our numerical experiments indicate that, on the one hand, a cross-shaped strategy is often more advantageous for urban expansion than a square shape. On the other hand, the “spatiotemporal” hybrid structure has great flexibility, and HTS-TM delivers outstanding performances in most scenarios tested (the benefit of HTS-TM can reach more than 10% compared with the state-of-the art transit systems). Even in low demand areas with a more or less uniform demand pattern, HTS-TM demonstrates that in the city periphery, a grid may not be worse than a hub-and-spoke network for transit.