On avoiding panic by pedestrian route recommendation in smart spaces

In this paper, we study the adaptability of pedestrian route recommender systems to panic behaviors in smart space networks. Due to environmental unsafe conditions, under panic, people may exhibit herding and stampeding behaviors. To achieve efficient and safe pedestrian transit, these behaviors should be prevented. Our approach to this issue is based on finding critical network areas where personally adaptable real-time smart space guidance will facilitate a coordinated people flow while reducing irrational behaviors. It is known that pedestrians tend to evacuate together if they are familiar with one another. Thus, we apply individual and collective rationality in pedestrian route optimization depending on the characteristics of each pedestrian. The adaptability of the proposed method to changes in structural and psychological circumstances regarding the proposed routes is demonstrated by means of a case study.