Recurrent emotional contagion for the crowd evacuation of a cyber-physical society

• Published in: Information sciences 2021-10, Vol.575, p.155-172
• Main Authors: Liu, Heng, Lu, Dianjie, Zhang, Guijuan, Hong, Xiao, Liu, Hong
• Format: Article
• Language: English
• Subjects: Crowd evacuation; Cyber-physical society; Recurrent emotional contagion; Small-world networks
• ISSN: 0020-0255; 1872-6291
• DOI: 10.1016/j.ins.2021.06.036

•We propose the CPS-REC model to describe the emotional contagion process of crowd evacuation more realistically.•We build the degree-based Mean-Field Equations for the CPS-REC model to describe the dynamic evolution of emotional contagion in CPS.•We construct the recurrent small-world networks, which is a special case of CPS-REC, to verify the validity of our model.•We implement a crowd evacuation simulation system to visualize the results of the CPS-REC model. In recent years, emotional contagion has drawn great attention in the field of public security since it often leads to stampedes and crushes during large-scale emergencies. To date, there have been many studies on emotional contagion in both cyberspace and physical space, bringing a strong impetus to the simulation of crowd evacuation in emergencies. However, due to the heterogeneity of crowds in cyberspace and physical space, emotional contagion is rarely comprehensively considered in the cyber-physical society (CPS). In addition, the characteristics of emotional recurrence prone to appear in large-scale evacuation have not yet been fully elucidated. Therefore, the existing models cannot accurately describe the real situation of emotional contagion since there is a wide deviation between it and the reality. To solve this problem, we propose a recurrent emotional contagion (REC) method for crowd evacuation of the CPS. First, we construct a CPS-oriented REC model by considering the influence of emotional recurrence on the emotional contagion process. Second, we build the degree-based Mean-Field Equations to describe the dynamic evolution of the number of individuals while considering the heterogeneity of crowds. Furthermore, we use the Finite Difference Method to derive the numerical solution. Third, we construct recurrent small-world networks, which constitute a special case of CPS-REC, to verify the effectiveness of our model. Finally, we implement a crowd simulation system based on the CPS-REC model to visualize the results of our theoretical analysis. The experimental results show that our method can more realistically simulate the emotional contagion process.