zapiano / vaxcounselors.jl

VaxCounselors.jl

VaxCounselors is a envy-free based model for vaccines distribution.

---

This is a model developed for a Master's Project at Universidade de São Paulo (University of São Paulo). It simulates the distribution of vaccines for a population of individuals in a pandemic context using an Envy-Free strategy. Other strategies (Maximize-Benefit, Minimize-Benefit, Oldest-First and Random-Vaccination) were also implemented for comparison purposes.

Quickstart

To install VaxCounselors.jl run

julia> ]
(my_project) pkg> add <path-to-package>

And to run all strategies for a country and some setups:

julia> using VaxCounselors

julia> countries::Vector{String} = ["usa"]
julia> setups::Vector{Int64} = [1,2,3,4]

julia> VaxCounselors.run_model(countries, setups)

Multithreading and Multiprocessing

To start julia enabling multithreading and multiple cores the flags --threads and --procs can be used .

$ julia --project=./ --threads=auto --procs=6

More info on that can be found in Julia's multithread page and multi-processing page.

vaxcounselors.jl's Issues

Fix wrong mean cumulative benefit plot

Change `sides` from "I" and "II" to symbols :L and :R

Save vaccines distributed for every time step

Only needed for envy-free strategy.

Bring plot to this project

Currently the plots are inside a Python jupyter notebook.

Change counselors "A" and "B" to symbols :A and :B

Fix Vaccinated Population plot yticks

Plot vax_population

Adjust all plots grid sizes and ticks

• Utilities
• Benefits
• Cum Benefits
• A-B
• A-B Cum
• Mean Cum Benefit
• Summary
• Vaccinated Population

Convert countries from Strings to Symbols

Save utility functions dataframe

Plot scatter + histogram for all countries, strategies and setups

Enable changing theme and apply seaborn nice looking theme

• Summary
• Utilitiy functions
• Benefit time series
• A-B time avg
• A-B cum time avg
• Mean cum benefit
• Vax population