• DocumentCode
    643329
  • Title

    An Optimal Transport-Based Approach for Crowd Evacuation

  • Author

    Kerrache, Said ; Benhidour, Hafida ; Al-Homaidhi, Sahar

  • Author_Institution
    Coll. of Comput. & Inf. Sci., King Saud Univ., Riyadh, Saudi Arabia
  • fYear
    2013
  • fDate
    24-25 Sept. 2013
  • Firstpage
    142
  • Lastpage
    147
  • Abstract
    The effectiveness of crowd evacuation is an important public safety issue. Crowd related incidents are frequent and often result in serious material and human loss, hence the increasing interest in developing new techniques to assist in crowd evacuation. In this paper, we introduce a new crowd evacuation method, where the evacuation plan is computed based on optimal transport. The use of the optimal transport formulation results in the minimization of the kinetic energy necessary to perform the evacuation, thus reducing the effort exercised by the crowd members. To ensure the safety and the feasibility of the computed plan, a safety model is introduced as an additional factor in the optimization. The proposed model allows avoiding dangerous places, spreading out the crowd, limiting the crowd density and imposing one-way circulation. A numerical method adapted to the resulting optimization problem is presented. The efficiency of the proposed approach is evaluated through simulation. The experimental results show that the proposed method computes a solution that strikes a balance between the different considered factors.
  • Keywords
    numerical analysis; occupational safety; optimisation; crowd density; crowd evacuation method; crowd members; kinetic energy; numerical method; one-way circulation; optimal transport formulation; optimal transport-based approach; optimization problem; public safety; safety model; Computational modeling; Kinetic energy; Mathematical model; Numerical models; Optimization; Safety; Vectors; crowd evacuation; crowd simulation; optimal transport; safety model;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computational Intelligence, Modelling and Simulation (CIMSim), 2013 Fifth International Conference on
  • Conference_Location
    Seoul
  • Print_ISBN
    978-1-4799-2308-3
  • Type

    conf

  • DOI
    10.1109/CIMSim.2013.31
  • Filename
    6663177