Introduction

In recent years, an important application for multi-agent models has been epidemiology: how to model contagion within a population. This module takes us to simulate the propagation of an infectious disease in a contact network of individuals. We combine our knowledge of graphs in a multi-agent context with our newly-gained knowledge of modelling time.

Learning outcomes

This module will help you do the following:

• Apply a generation model for complex networks (specifically scale-free networks)
• Model a propagation in a network (specifically a SIR model)
• Choose pseudorandom subsets
• Define a multi-stage simulation step
• Record step-by-step data of interest in the form of time series from a simulation (without having to store or visualize the full details of each step)
• Define a stopping condition (more sophisticated than a maximum step count)
• Visualize time series

Warm-up

First, read the BBC news story on how contagion rates are quantified. Then, play for with the Washington Post Corona Simulator for a few minutes to get a general gist of contagion modeling.

Warm-up assessment

Having now studied, briefly, epidemiology, discuss on what aspects of society you find essential to incorporate in a simulation of an outbreak of a contagious disease in order for that simulation to be realistic enough to produce actionable insights yet abstract enough to be fast and easy to implement and study. List the aspects and, for each, give your suggestion on what and how to model; for example, if the aspect were "commute", one could simulate which individuals will be enclosed in a shared vehicle, how often, at what degree of proximity, and for how long at a time.

Concepts

After this module, you should be familiar with the following concepts:

• Generation model
• Preferential attachment
• SIR model
• Time series

Remember that you can always look concepts up in the glossary. Should anything be missing or insufficient, please report it.