Análise do comportamento da Covid-19 no Brasil e em Sorocaba via modelagem matemática epidemiológica

Abstract

Mathematical modeling of epidemiological phenomena, using systems of differential equations, combined with the evolution of computational techniques, increasingly contributes to decision-making processes, prevention and control of diseases. This Course Completion Work has as its theme studies on the Covid-19 pandemic in Brazil and in the city of Sorocaba/SP, making use of the compartmental mathematical model SIR (Susceptible, Infected and Removed), which was solved numerically via Euler's method. The objectives consisted of running simulations of the model, in Python language, collecting real Covid-19 data in the country and in the municipality, estimating the system parameters and plotting graphs capable of helping to understand the dynamics of the pandemic. Data regarding Brazil were taken from the Coronavirus Brasil digital platform, made available by the Federal Government in partnership with the country's health units. The information regarding the municipality of Sorocaba was collected on the SP Covid-19 Info Tracker digital platform, prepared by a committee of researchers from Unesp, USP and CeMai, which centralized the records in daily Covid-19 bulletins, taking into account the data provided by the municipalities of the state of São Paulo. This research began in a scientific initiation project, developed between 2020 and 2021 and arose from the great need for knowledge about the spread of the disease across the country and the search for alternatives that were appropriate for making satisfactory predictions, with a minimum of resources possible. The results achieved showed the importance of an adequate adjustment for the disease transmission parameter and confirmed that it is possible for a relatively simple SIR model to produce solutions in the same order of magnitude as the real data, serving as a forecasting tool to a certain extent of the pandemic. In addition, a model like this allows validating the application of public measures, verifying whether there has been an increase or decrease in the number of cases, in relation to what was expected.