doi:10.3808/jei.200800125
Copyright © 2021 ISEIS. All rights reserved

TISEM: A Two-Stage Interval-Stochastic Evacuation Management Model

G. C. Li1,G. H. Huang1*,C. Z. Wu2,Y. P. Li3,Y. M. Chen1,Q. Tan1

  1. Faculty of Engineering, University of Regina, Regina, Saskathewan S4S 0A2, Canada
  2. Intelligent Transportation System Research Center, Wuhan University of Technology, Wuhan 430070, P. R. China
  3. College of Urban and Environmental Sciences, Peking University, Beijing 100871, P. R. China

*Corresponding author. +1-306-5854095 Fax: +1-306-5854855 Email: gordon.huang@uregina.ca

Abstract


Traffic allocation planning is commonly required for mass evacuation management. It primarily relies on efficient coordination and appropriate utilization of roadway capacity and available traffic resources. However, traffic and evacuee information are usually difficult to be obtained and consequently of various uncertainties in data. Especially, stochastic information may often exist in evacuation management systems. In this study, a two-stage interval-stochastic evacuation management (TISEM) model was developed for supporting the evacuation planning under uncertainty, by which stochastic and interval evacuation information could be well reflected and communicated in the system. In addition, by adopting the proposed model, a case study abstracted from the City of Wuhan was introduced and solved through an interactive method. Results indicated that useful solutions for planning evacuation routes could be generated based on results of the model. As well, through the model, complex relationships between evacuation time, environmental influences and economic factors could be systematically analyzed. It demonstrated that the proposed TISP model is practical and applicable in real world, and is helpful for authorities to make decisions allocating vehicles before evacuation starts.

Keywords: allocation planning, management, optimization, stadium evacuation, transportation, uncertainty


Full Text:

PDF

Supplementary Files:

Refbacks

  • There are currently no refbacks.