Multi-Agent Simulations are useful tools to predict the effects of public policies. In the last three years, with the concerns around the COVID-19 pandemic, several simulations were developed to understand the effects of lockdown, travel, etc. Even before that, MAS systems were used to plan disaster evacuation policies, transit policies, and many others. In this paper, we propose and analyze a mixed model that considers the effects of masking and large scale evacuations at the scale of a large university campus and its neighborhood. This project is part of a larger effort to create a simulator that considers how human mobility (pedestrian, public transportation, private transportation) interacts with large scale events (natural disasters, entrance examinations, pandemics) at a neighborhood level in the Japanese context. We evaluate how the simulator in its current state can reflect the effect of different masking policies on the spread of COVID-19 during an earthquake evacuation scenario.

Simulating the development of cities is interesting from the point of view of understanding human communities, but also brings benefits to city planners. However, understanding the expansion of land use and transportation networks is a known challenge. In this work, we investigate a city development model that combines a rule-based procedural road generation algorithm with multi-agent simulation of land choice and movement through the map using the generated land use and transportation network. The city map is organized in a grid, and an initial land value for each cell is calculated based on its geospatial features. Next, a set of agents are randomly initialized and perform actions on the city map, such as establishing residences, commuting, and trading in the exploration phase. Then, the actions performed by the agents are used as parameters for recalculating land prices and guiding the expansion of the road network, in a network development phase. We evaluate the emergence of geometrical patterns in the road network as well as land use and population distribution in the final map. We also compare maps generated using geographical data from selected locations to their corresponding real world settlements.