EGAR: Environment Generator for Agent-Based Research

Abstract

[Abstract]: Agent-based modeling (ABM) allows the representation of real-world phenomena using agents, a virtual environment, and interactions between them. It has become one of the most important tools for studying complex phenomena in social sciences.

The correct placement of agents, locations, structures, and objects in the virtual environment, e.g. a city, is important in the development of an ABM in order to increase the realism and accuracy of the simulations.

Unity is a game engine that allows the development of ABM with the advantage of greatly optimizing the management of resources compared to other alternatives, making it possible to reduce simulation time or handle more complex models. In this work, a tool is proposed that allows the generation of a more precise virtual environment for the development of ABM in Unity.

The tool developed facilitates the visualization and distribution of agents and the creation of networks between them. It provides a way to explore and understand the social and spatial processes taking place in the simulated environment.

By combining ABM, GIS data, and visualization tools in virtual environments, researchers have a robust framework for tackling complex modeling problems. For example, in social sciences, this integrated approach allows the exploration of hypothetical scenarios, the evaluation of intervention strategies, and a better understanding of networks of influence. Two real-world problems, the spread of a virus and the implementation of a new urban plan based on superblocks, are presented to illustrate this tool.