Collective patterns of pedestrians interacting with attractions

Walking is a fundamental activity of a human life, not only for moving between places but also for interacting with surrounding environments. While walking to destinations, pedestrians may acquaint themselves with attractions such as artworks, shop displays, and public events. If such attractions are tempting enough, pedestrians opt to stop walking to join the attractions. Although the existence of attractions may considerably affect the pedestrian flow patterns, little attention has been paid to the interactions between pedestrians and attractions, and their impacts on pedestrian traffic. This dissertation developed two microscopic models of pedestrian flow interacting with attractions. In numerical simulations, the presented models were examined by systematically controlling model parameters. After performing the numerical simulations, various collective patterns were identified and summarized in phase diagrams based on macroscopic measures. By doing so, this dissertation investigated the dynamics of pedestrian flow interacting with attractions from the perspective of attracted pedestrians and passersby. The first model represents the attractive force towards the attractions in line with the social force models. The first model predicted various collective behaviors of attracted pedestrians, such as forming stable clusters around the attractions and rushing into the attractions. To understand collective patterns of pedestrians' visiting behavior, the second model was formulated as a function of the social influence and the average duration of visiting an attraction. The second model suggested that increasing the social influence and the average duration of visiting an attraction tended to attract more visitors to the attraction, but the increment was effective for a certain range of the parameters. Based on the second model, jamming transitions in pedestrian flow interacting with an attraction were also studied. It was found that an attendee cluster can trigger jamming transitions by increasing conflicts among pedestrians near the attraction. This dissertation contributes to the current body of knowledge on the collective behavior of pedestrian motions. This is achieved by modeling pedestrian dynamics interacting with attractions and providing possible explanations of the collective patterns. The presented models and their applications enable one to understand various collective patterns of pedestrians interacting with attractions. The findings presented in this dissertation can provide an insight into pedestrian flow patterns in stores and improve the understanding of collective phenomena relevant to pedestrian facility management.