Horizon CDT Research Highlights

Research Highlights

Spatial choice and spatial behavior of Mega-city Public Transportation System Driven by Multi-source Data

  Shuhui Gong (2015 cohort)   www.nottingham.ac.uk/~psxsg7


With the development of transportation, city could be much broader than before, even very close to suburb. In mega-city in China, many people reside in suburb, and commute from suburb to city center. Some mega-city in China, like Beijing, Shanghai, and Shenzhen has their own commute zone, where citizens live in, and go to city center to work every day. To some extent, if the business in city spread very fast even to remote area, then new business center will be set up outside the city, then people who live in city center may need to commute to suburb to work. Commute may cause serious problems to mega-city, especially increasing the heavy of public transport.[1]

As it is widely known, it only costs 26 years to Shenzhen from a small fishing village to one of the most prosperous city in China, which surely must contribute to the designation of “Special Economic Zone” in 1908s. During this time, it has exceeding 8.5 million people.[2] The rapidly developing business brings Shenzhen a serious “commute zone”, then how to use commute model to solve the particular problem to help reduce the force of the transport in city is a real challenge.

The purpose of the project includes three parts: First is description: research need to start to set up a new spatial choice model or modified an existing model to describe the situation of Shenzhen transport, the main commute problem which may related to time and distance. Second is discipline: after the mega-city transportation can be described by a relatively accurate model, data mining should be used to find the discipline of the transport, such as the inside relationship between human behavior and transport congestion. Third is optimization: after finding the potential discipline of the transport factors, the project will focus on one typical problem, use spatial choice and behavior model to help solve it. The expectation of the optimization is giving suggestion on the design of the transport, which may improve the existing transport environment in Shenzhen.


  • Gravity model.
  • Minimum Spanning Tree (MST)
  • The universal model of commuting network
  • Multi-scale Mobility Networks


  1. J, O.; L, W. Modeling Transport. New York: John Wiley and Sons Ltd. 2011.
  2. LIU, P. et al. Coastal city subsidence in Shenzhen (China), monitored using multi-frequency radar interferometry time-series techniques 2014.


  1. Zhiyuan Xu, Shuhui Gong, Zhaojing Zhang, Bin Jia, Hao Jin, Dong Liang: Effects of the Tradeoff between the Customer Churn Rate and Retention Capability on the Predictive Performance of Algorithms. ICDM (Posters) 2015: 10-16

This work was carried out at the International Doctoral Innovation Centre (IDIC). The authors acknowledge the financial support from Ningbo Education Bureau, Ningbo Science and Technology Bureau, China's MOST, the University of Nottingham, and Shenzhen key laboratory of spatial smart sensing and services, Shenzhen University. The work is also partially supported by EPSRC grant no EP/L015463/1.