Predictive Modeling of Pedestrian Behavior in Mass Gatherings: a Study of Kumbh Mela

Organization of the mass gathering that would be both fulfilling its entertainment, cultural or religious purpose and safe for all of its visitors often requires substantial preparation from the side of organizers. Kumbh Mela - a religious festival that beings together around 100 millions of devotees is one of the largest mass gatherings in the world. An international research team of scientists from ITMO University has conducted a set of experiments at the site of the festival - in Ujjain,  India in May 2016. Given the obtained data, the development of technologies, mathematical models, methods and algorithms aimed at assisting complex security of global mass gatherings has been planned.

Human crowds are complex systems and predicting their behavior, hence, is a complex task. Nevertheless, combination of computer modeling and data analysis methods reveals approaches that can help manage such events. Reports about similar events, aggregate variables calculations can be used as a primary data source for supporting  the decisions made by organizers. With the advent of computational technologies, organizers of mass gatherings have been offered new methods - models of crowd movement that allow to reproduce different scenarios that may take place during such events even in cases where precedent is yet nonexistent.


  • Karbovskii V.A., Voloshin Daniil V., Karsakov A., Bezgodov Alexey, Zagarskikh A.

    Multiscale agent-based simulation in large city areas: emergency evacuation use case // Procedia Computer Science. — 2015. — Vol. 51. — pp. 2367-2376.

  • Voloshin D.V., Rybokonenko D., Karbovskii V.A.

    Optimization-based Calibration for Micro-level Agent-based Simulation of Pedestrian Behavior in Public Spaces // Procedia Computer Science. — 2015. — Vol. 66. — pp. 372-381.


  • ФЦП ИР №14.584.21.0015 от 11.11.2015 г.