A study on the bottleneck effect observed in an emergency evacuation exit employed by multi-agent simulation and mean-field approximation analysis

An improved cellular automaton model based on Yanagjsawa & Tomoeda , Nishinari (2007) was established, where both Static Floor Field and collision effect were considered Several model parameters were carefully determined by going through a turning process based on experimental data provided by other previous studies. Both results by simulation based on the model and analytical approach derived from the so-called Mean-Field Approximation proved that the outflow rate from an evacuation exit, usually estimated by the so-called flow coefficient, can be improved by putting appropriate obstacles in front of the exit This is because the appropriate allocation of obstacles can deflate collision probability at the exit by increasing collisions around the obstacles before the exit.