David Gabriel de Barros Franco

All public administration contracts must be evaluated in order to improve their performance while respecting the limits established by laws and regulations. The purpose of this article is to apply an approach to improve the performance of a common building maintenance service contract for the Paraná State Court of Justice (PRCJ), which has a total built area of ​​544 283,79 m² in 224 buildings distributed over 161 counties, by optimizing the territorial partitioning of Paraná State. To partition the state into 14 regions, a binary integer linear programming (BILP) mathematical model is applied to the facilities location problem (FLP) in three scenarios. The results show that Scenario 3 (in which the location of the 14 maintenance offices and the configuration of their areas of activity were optimized) is the best in terms of minimizing the distances traveled by maintenance teams. In this scenario, the total distance traveled would be 9 775 km per day (instead of the current 11 150 km), achieving savings of around 12,3% when compared to the current solution. With this solution, in addition to the distance, the direct and indirect costs associated with the displacement of work teams and the time spent on their corresponding trips would be minimized. Furthermore, the users of maintenance services could be served more quickly, resulting in a higher number of services and greater satisfaction for the target audience of the contract.

RESUMO Neste trabalho foram utilizados, conjuntamente, os conceitos de particionamento territorial e localização de instalações para a otimização da rede de aterros sanitários do Estado do Paraná. Isso porque a solução adotada atualmente apresenta uma série de deficiências, incluindo o não atendimento total da demanda, o número excessivo de pequenos locais e diversos consórcios não contíguos, que resultam em maiores custos de transporte e menor eficiência de escala. O objetivo desta pesquisa é otimizar o particionamento territorial para o Estado do Paraná com relação aos aterros sanitários definindo um número adequado de lugares capaz de atender à demanda e que minimize as distâncias a serem percorridas dos pontos de geração do lixo até o descarte. Para isso, foram desenvolvidas simulações em função do custo de transporte rodoviário dos resíduos, em que foram considerados os locais dos 171 aterros operacionais existentes no Estado. Os resultados mostraram a necessidade de se alterarem algumas das capacidades das áreas de descarte existentes, além de se adicionarem 25 novas áreas para atender integralmente à demanda. Com isso, o gasto com transporte seria reduzido em cerca de 80% e o custo total, que inclui novas construções, ficaria reduzido em 57%, acarretando redução de gastos, além do aprimoramento na prestação do serviço e da atenuação do impacto ambiental.

During the pandemic caused by the Coronavirus (Covid-19), Machine Learning (ML) techniques can be used, among other alternatives, to detect the virus in its early stages, which would aid a fast recovery and help to ease the pressure on healthcare systems. In this study, we present a Systematic Literature Review (SLR) and a Bibliometric Analysis of ML technique applications in the Covid-19 pandemic, from January 2020 to June 2021, identifying possible unexplored gaps. In the SLR, the 117 most cited papers published during the period were analyzed and divided into four categories: 22 articles that analyzed the problem of the disease using ML techniques in an X-Ray (XR) analysis and Computed Tomography (CT) of the lungs of infected patients; 13 articles that studied the problem by addressing social network tools using ML techniques; 44 articles directly used ML techniques in forecasting problems; and 38 articles that applied ML techniques for general issues regarding the disease. The gap identified in the literature had to do with the use of ML techniques when analyzing the relationship between the human genotype and susceptibility to Covid-19 or the severity of the infection, a subject that has begun to be explored in the scientific community.

This work aimed to apply the ARIMA model to predict the under-reporting of new Hansen’s disease cases during the COVID-19 pandemic in Palmas, Tocantins, Brazil. This is an ecological time series study of Hansen’s disease indicators in the city of Palmas between 2001 and 2020 using the autoregressive integrated moving averages method. Data from the Notifiable Injuries Information System and population estimates from the Brazilian Institute of Geography and Statistics were collected. A total of 7035 new reported cases of Hansen’s disease were analyzed. The ARIMA model (4,0,3) presented the lowest values for the two tested information criteria and was the one that best fit the data, as AIC = 431.30 and BIC = 462.28, using a statistical significance level of 0.05 and showing the differences between the predicted values and those recorded in the notifications, indicating a large number of under-reporting of Hansen’s disease new cases during the period from April to December 2020. The ARIMA model reported that 177% of new cases of Hansen’s disease were not reported in Palmas during the period of the COVID-19 pandemic in 2020. This study shows the need for the municipal control program to undertake immediate actions in terms of actively searching for cases and reducing their hidden prevalence.

In this paper we proposed two new strategies for initialization and training of Radial Basis Function (RBF) Neural Network. The first approach takes into consideration the "error" between the input vector p of the network and the x-axis, which are the centers of radial functions. The second approach takes into account the "error" between the input vector p and the network output. In order to check the performances of these strategies, we used Brazilian financial market data for the RBF networks training, specifically the adjusted prices of the 10 greater weighted shares in the Bovespa index at the time of data collection - from April 8^th, 2009 to October 31^th, 2014. The first approach presented a 52% of improvement in the mean squared error (MSE) compared to the standard RBF network, while the improvement for the second approach was 38%. The strategies proved to be consistent for the time series tested, in addition to having a low computational cost. It is proposed that these strategies be improved by testing them with the Levenberg-Marquardt algorithm.

Este trabalho visa comparar Redes Neurais Artificiais (RNAs), especificamente o Perceptron de Múltiplas Camadas, a Rede Alimentada Adiante Focada Atrasada no T...