Discret-event-virus-simulation

( the report is the result part )

installation

go in your bash and execute

pip3 install simpy
pip3 install matplotlib.pyplot
pip3 install numpy

Version : python 3.6.9 64bits

simpy

what is simpy ?. we used simpy in order to make a discrete event simulation.

config file

config:
    RANDOM_SEED : 1111
    NUM_AREA : 2                         # number of meeting zone (2 persons can be enter is this zone)
    TIMEMEET : 10                        # minutes staying in the meeting zone
    SIM_TIME : 20000                     # Simulation time in minutes (in one cycle)
    NUM_PERSON : 10000                   # number of person in the simulation
    NUM_TIPS : 3000                      # number of object (person) go in meeting zone (in one cycle)
    NUM_CYCLE_OUTPUT : 20                # number of cycle
    P : 1                                # probabilty to infect someone

start my simulation

run this command in your shell in order to start the simulation

python3 simulation.py > log.txt 

this command allows you to write the application log in a txt file.

run this command in your shell in order to visualise the simulation.

python3 analyse.py 

analyse.py process the file log.txt

resulats

without containment

We observe that the number of infected people is increasing exponentially. The peak of infected people (blue curve) is reached with 1500 cases of contamination in only 1 day for a population of 10 000 inhabitants. Towards the end of the epidemic, the red curve becomes constant because the entire population has been infected.

let's go into more detail

I'm doing a linear regression from the first quartile to the third quartile. We focus on the coeficient of the slope. without the containment this coeficient is ~ 926. which means there are 965 more people infected every day.

with containment

to simulate containment, it is assumed that the number of people exiting is reduced by 90%.

so we change the config file :

config:
    NUM_TIPS : 3000                      # number of object (person) go in meeting zone (in one cycle)
    NUM_CYCLE_OUTPUT : 120               # number of cycle
    P : 0.8                              # you have less chance to infect someone because your are in a containment

It takes longer for the virus to spread. It is only after the 40th day that the virus will start infecting many people. The virus affects entirely the population from the 100th day when previously we saw that the virus contaminates all the population in only 17 days !

In addition, the peak of infected / days is only 590

Linear regression allows us to say that containment reduces the number of infected people per day.

I can therefore conclude that the confinement allows us to reduce the congestion in hospitals and to have a less brutal epidemic.

other

Here is the link to my non-discrete event simulation project: https://github.com/Debzou/SimulationVirusContainment