Jamming transitions induced by an attraction in pedestrian flow

We numerically study jamming transitions in pedestrian flow interacting with an attraction, mostly based on the social force model for pedestrians who can join the attraction. We formulate the joining probability as a function of social influence from others, reflecting that individual choice behavior is likely influenced by others. By controlling pedestrian influx and the social influence parameter, we identify various pedestrian flow patterns. For bidirectional flow scenario, we observe a transition from free flow to freezing phases in which oppositely walking pedestrians reach a complete stop and block each other. On the other hand, a different transition behavior appear in the unidirectional flow scenario, i.e., from free flow to localized jam, and then to extended jam phases. It is also observed that the extended jam phase can turn into freezing phase with a certain probability when pedestrian flux is high with strong social influence. We suggest that an attendee cluster can trigger jamming transitions not only by reducing the available walking space but also by increasing conflicts among pedestrians near the attraction.