High-speed rail networks, capacity investments and social welfare

Valentina Bracaglia, Tiziana D'Alfonso, Alberto Nastasi, Dian Sheng, Yulai Wan, Aming Zhang

Research output: Journal article publicationJournal articleAcademic researchpeer-review

14 Citations (Scopus)


In this paper, we analytically study the performance of two topologies of high-speed rail (HSR) networks: isolated-corridors and grid networks. We evaluate how HSR configuration affects capacity of newly developed infrastructure, profits and social welfare by considering a number of factors, namely economies of traffic density, market size, operating cost and cost of capital. Our investigations focus on a social welfare-maximizing entity which provides HSR train services as well as develops infrastructure. We find that although the grid network allows for more new markets, the isolated-corridors network may perform better in terms of social welfare when the cost of capital is high, the demand in the new markets established by the isolated-corridors network is high, the operating cost (excluding density effect) is high and the traffic density effect is weak. We also identify cases where the optimal network configuration in terms of social welfare may not be optimal in other aspects, such as capacity, consumer surplus and profits. For example, when the density effect is strong, grid network is likely to be socially optimal, but it faces more difficulty in recovering the invest cost comparing with the alterative network.

Original languageEnglish
Pages (from-to)308-323
Number of pages16
JournalTransportation Research Part A: Policy and Practice
Publication statusPublished - Feb 2020


  • Capacity
  • High-speed rail
  • Network configuration
  • Social welfare
  • Traffic density effect

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Transportation
  • Management Science and Operations Research


Dive into the research topics of 'High-speed rail networks, capacity investments and social welfare'. Together they form a unique fingerprint.

Cite this