Gunderfish Logo
This model is for educational purposes only. If you would like more information, contact us.

Introduction to the GunderFish COVID-19 Simulation

It is April of 2020, and the world is dealing with a pandemic caused by the SARS-COV-2 virus, and the related disease COVID-19. One question on almost everybody's mind is "How will this disease spread?" At GunderFish we are helping answer that question.

The model presented here is a simplified version of the types of models being used to predict the spread around the world. It is designed to provide a clear, visual demonstration of the effects of some of the social distancing and quarantine policies on controlling the disease.

Nurse in scrubs and mask looking out from a temporary tent hung with signs for 'Covid-19 testing' in English and Spanish
Fig.1 COVID-19 Testing Station in New York City on 31Mar2020.

The model is based on tracing the progression of the disease through a population, and applying various response policies at specific times during the outbreak. We track the initially uninfected population as they come in contact with infected people, get infected, become symptomatic, and finally to a resolution state: either recovered and (now) immune to the disease, or to a more fatal outcome.

Along the way, we look at decisions made about isolating individuals - these could be people deciding to isolate themselves, regardless of whether they are infected, people becomimng symptomatic and either self isolating or being quarantined; and people becoming sick enough to require hospitalization and becoming isolated in the facility.

The model uses three main components:

  1. A model of the disease
    How infectious is it?
    How long symptoms take to appear?
    How long does it take to get over it?
    How many of the infected will survive?
  2. A model of the population
    How many people are in the area
    How socially active they are (do they have close contact with 1 or two people a day, or dozens?)
    How socially responsible are they (do they go back to work even if they are sick, or do they voluntarily stay home)
    Other aspects of their culture.
  3. A model of the types of control that the government can apply.
    Can they force everyone to stay home for 2 months, or are they only able to ask people to avoid contact.
    Can they trace the contacts of every infected person?
    Can they produce millions of tests quickly, or will it take time?

Gunderfish Outbreak Simulator State model
Fig.2 Population states and transitions. Each day the portion of the population in any bubble has a probability of moving to an ajacent bubble. The probability is determined by the nature of the disease, and the types of control policies in effect at the time.

These three components allow the model to estimate how quickly the virus will spread through he population; how many people will get sick and how quickly; and what the government can do to control the outbreak. The model flow is shown above.

Next Steps

Check out the Quick Start guide, and see how well you can fight the pandemic!
Image of the main control panel for the GunderFish COVID-19 Simulator
Fig.3 - Main Control Panel GunderFish Simulator.