Disclaimer

This SEIRD model is put together for a school project and for general interest. If you want an accurate model that models out COVID-19 and other pandemics for reasons other than general interest, please seek elsewhere. As of April 2020, this project is considered complete and archived.

SEIRD Model

Stages

![Susceptible -> Exposed -> Infectious -> Dead/Recovered][https://github.com/cjinn/seirdp-model/blob/master/SEIRD.png]

This dynamic, heuristic, epidermic model consist of six stages:

1. Susceptible - Population that is susceptible to a disease
2. Exposed - Population that is exposed to the disease but do not show symptoms yet
3. Infectious - Population that is infectious and spreads the disease to others
4. Recovered - Population that has recovered and can no longer spread the disease
5. Dead - Population dead by the disease

This model consist of ordinary, differential equations that attempt to model any epidermic (such as COVID-19). The many parameters included in here are to help make different inferences on what parameters are important to a pandemic response.

Assumptions

• Population is a closed population (but it may grow or decline)
• Population is not immune to the disease; given the chance, everyone will become infectious and may die
• Many variables are reduced down into simple, numeric constant rates. Many do not change with time
• Model is heuristic and deterministic (no randomness)
• People who show symptoms and people are asymptomic but are still infectious are lumped together into 'Infectious' stage
• Population in 'Recovered' stage do not become susceptible to the disease
• There is no natural or induced immunity
• People who die stay dead and are no longer infectious

Routes

There are two main routes for everyone in the susceptible population to take:

• Susceptible -> Exposed -> Infectious -> Recovered
• Susceptible -> Exposed -> Infectious -> Dead

Running my Code

Installation

1. Install Python3
2. Run the command pip3 install -r requirements.txt

Demo (Python Backend)

1. Run the command python3 seird.py

Demo (GUI)

1. Run the command python3 userinterface.py
2. Enter parameters as you see fit
3. Click 'Ok' and a plot should appear

Note that this demo runs off covid_params.py. These numbers are hypothetical numbers.

Parameters

There are many parameters when modelling a disease. This model attempts to account for many different scenarios. See covid_params.py to get a good feeling of what parameters are used.

Here is the list of parameters that you should change:

• r0 - Basic Reproductive Number of the disease (unrestrictive)
• rc - Basic Reproductive Number of the disease when social distancing is implemented
• gamma - The rate an infectious person recovers and moves into the recovered phase. Note that this means they do not infect anybody any more.
• sigma - The rate at which an exposed person becomes infectious. This is defined as 1/(incubationPeriod)
• baseAlpha - Probability that the disease will kill a person
• rho - Rate at which people die (1/6 = 6 days to kill a person)
• socDistResponseFactor - Population's receptiveness to social distancing. Range at [0, 1]. Defaults at 1.0. The higher this number is, the likelihood the population responds positively to the countermeasures.
• diseaseScalingFactor - How more deadly the disease is the greater the population of the infectious. Range at [0, 1]. Defaults at 0.0. The more overworked the system is, the higher this number gets.
• population - Population number. Note that population system is closed.
• E0 - Initial seed amount of people infected with the disease.
• thrDay - Day that triggers social distancing. Set this higher than daysModel to never trigger it.
• daysModel - Number of days to simulate the model. 150 days provides good resolution for the figure.

References

• https://towardsdatascience.com/infectious-disease-modelling-beyond-the-basic-sir-model-216369c584c4
• https://github.com/coronafighter/coronaSEIR/blob/master/main_coronaSEIR.py