COVID 19 Dashboard/Simulation

• Visualize cases of COVID-19 virus around the world and compare them with simulation data.
• Iterative multi-level simulation meant to be run on the country scale. The behavior of each person is simulated on a daily basis using probabilistic data.
• Compare with real data to analyze how the measures implemented to prevent the virus spread are affecting its evolution.

• GUI can be used to test simulation configurations and compare results with real-world data from all countries.
• Its possible to use countries data to automatically setup a simulation configuration.
• Simulation, country and Covid-19 data can be stored and loaded from JSON files allowing for offline simulation using node.

Headless Simulation

• Trying to run long tasks with high memory usage its pretty much impossible. To overcome this issue its possible to run simulation using NodeJS.
• Download the repository from GitHub, install NodeJS and install all dependencies from NPM. Run the headless simulation using npm run node.
• JSON files produced by the program can be loaded into the Web interface for inspection.

Simulation

• Simulation is composed by 7 logic elements, Country, District, Home, Person, Foreigners, Hospital and Governor. The country contains Districts that contain Homes with groups of Person.
• Each person movement behavior inside of its district and to the outside is simulated every day. Persons that live in the same home have contact with each other everyday.
• The governor is an outside looker. That analyses data every X days and activate measure to reduce contact contamination (e.g block movement to surrounding districts, reduce allowed foreigners in the country, etc). Allow to compare the measures adopted by different countries, and predict how the virus is progressing on a daily basis.
• Foreigners enter the country everyday they visit random districts and may be infected (with out without symptoms with the disease).
• Hospital take care of Persons they have a bed limit and a effectiveness ratio that reduces the chance of the Person dying / increases change of recovery.
• In the simulation model recovered Persons are (a ton) more resistant to the virus but they can infected again with really small chance.
• The simulation configuration is based on real data from countries including the population age and district count (based on city count). The simulation does not consider gender of the Person being simulated.

Data sources

License

• The project is distributed under MIT license available on the github page.