Individual-based distributed epidemic spread model

This paper applied the theory of distributed computing and designed an individual-based distributed epidemic spread model according to the traditional SEIRS model. In the proposed spread model, microcosmic individuals interact and infect with each other through the spatial contact and social relationship contact, and influence the macrocosmic epidemic spread. The whole region is separated into the same regular sub-regions with quadtree coding, and the social relationship is classified into different types. Individuals were classified into corresponding groups according to the spatial locations or the social relationship types. The final region combination was implemented with quadtree decoding. The processes of interactions between different sub-regions and relationship types were emphasized. Every single step of distributed computing included the reseparation and re-combination of sub-regions, and the boundary process of sub-regions. The simulation experiment showed the proposed distributed model could save 80% and 50% runtime in spatial spread and relationship spread than single computing model approximately.