An agent based modeling approach to evaluate crowd movement strategies and density at bathing areas during Kumbh Mela-2019

Kumbh-Mela of Prayagraj, India, a festival of faith and belief, is one of the many significant gathering events worldwide. Pilgrims arrive from different places to take a holy bath at the confluence point of 3-rivers the Ganges, Yamuna, and Sarasvati. The police department is assigned a major role of handling and managing the dense traffic of pilgrims in this event to avoid unwanted situations. The primary surveillance points are the intersecting junctions and bathing zones. In addition, the authorities make crowd movement plans with different route diversion schemes and sets bathing time intervals to maintain crowd density at the Kumbh Mela site. Significantly, we must test these crowd management plans for a realistic assessment of population count, density maintenance, and time management. In this paper, we have created a model utilizing a micro-modeling agent-based approach that incorporates the virtual environment of the site. We have used AnyLogic, a ABMS tool, to incorporate social forces and stochastic behavior among the synthetic agents. The model simulates different crowd movement plans according to real behavioral scenarios. In the simulation, we have considered the whole bathing procedure as a halt time in the area. We have utilized our model to evaluate the time consumed by the pilgrims to reach “Ghat”. Also, to count the number of pilgrims that took a bath in 12 hours on different time intervals set for bathing. The significance of performing these evaluations is to assess the effect on density at the site during the whole arrival, bathing, and departure process.