Public Transit Passenger Quantity Forecast Based on Improved Multi-Variable Grey Self-Adaptive Model

The research of urban public transit passenger quantity possesses important value for the rapid development of public transport and the effective relief of urban traffic pressure. The passenger transport system is a multi-level and nonlinear complicated system affected by multi-factors. Because the relationship between urban public transit passenger flow and the affecting factors has grey characteristics, and the existing statistics data are of short and incomplete, the authors put forward an improved grey forecast model to predict urban public transit passenger quantity. First, according to the calculation of grey incidence, the key factors, called forecast variables, of urban public transit passenger transportation quantity are discovered. These factors are the region total output value and the amount of service buses. Next, on the basis of the grey incidence result, they construct the improved multi-variable grey self-adaptive MGM (1, n) model. They increase the forecast value to the known sequence on by one; meanwhile, replacing individually the old data with relative large forecast error. They fill vacancy in proper order and replace in this way until achieving the forecast goal. Finally, they employ the model in public transit passenger quantity forecast in Shijiazhuang city, Hebei province, and obtain good forecast results. The case study indicates that the improved model can not only reflect the mutual influence and restrict relations of multi-variables in the passenger transport system, but also overcome the traditional forecasting method's shortage failing to respond to the outside influence factors. The error between the forecast results and the real one is small; therefore, the improved model has good application value in the passenger quantity forecast of public transit.


  • English

Media Info

  • Media Type: Web
  • Features: References;
  • Pagination: pp 4348-4353
  • Monograph Title: International Conference on Transportation Engineering 2009

Subject/Index Terms

Filing Info

  • Accession Number: 01535854
  • Record Type: Publication
  • ISBN: 9780784410394
  • Files: TRIS, ASCE
  • Created Date: Nov 12 2013 1:46PM