An interacting multiple model (IMM) filter is a recognized method for adaptive estimation of states that is often necessary to characterize the behavior of dynamic systems with a multiple-mode operation. The traditional IMM filter adopts the measurement set to update the information about the active models. However, when this approach is used for fault-prognostic applications, it can lead to misleading state estimation. To overcome this problem, this paper proposes a novel fault-prognostic technique based on a modified IMM filter with fuzzy systems. The main methodological contribution of this paper is to build an approach based on an IMM filter able to make long-term fault predictions without measurements. This is done by incorporating a fuzzy system into the proposed IMM filter with the objective of modeling the system's dynamics and updating the probabilities of the observed modes. Furthermore, the proposed method and the standard IMM filter are compared in a numerical example and a real experimental platform PRONOSTIA for validation. The results analysis indicates a better prediction performance than the conventional IMM filter considering the failure time predictions and a measurement model not properly adjusted.

In the forestry sector, efficient planning of activities is considered essential for companies to achieve their goals and remain competitive in the market. In order to assist in this process, this work proposes the development and evaluation of a heuristic for generation of valid solutions for a forest production planning problem involving 120 stands and 81 management alternatives per field, totaling 9720 decision variables. In the tests carried out, such proposed heuristic presented promising results when compared to literature, indicating that it is capable of providing good quality solutions with low computational cost. Results also indicate that the inclusion of solutions obtained by using the proposed heuristic in the set of initial individuals/antibodies of Genetic Algorithm and Clonal Selection Algorithm can generate a significant improvement in the quality of final solutions and, consequently, in the performance of such meta-heuristics.

Os sistemas industriais e seus componentes podem ser caracterizados por diferentes modos de funcionamento e falhas, principalmente porque várias partes podem sofrer com diferentes tipos de deteriorações. Por isso, saber o melhor momento para o início das atividades de manutenção pode prolongar a vida útil dos equipamentos. Para auxiliar a execução de manutenção desses equipamentos, o presente trabalho propõe uma abordagem de estimação da condição de funcionamento, utilizando uma estrutura variável para os filtros de partículas com múltiplos modelos, aplicado à estimação do estado de uma bateria de íon-lítio. Duas abordagens são propostas (IMMA e IMMAND), diferindo na lógica de seleção dos modelos ativos mais prováveis. Para avaliar e validar os resultados obtidos pelas abordagens, são utilizados as métricas RMSE e MAPE, que mostraram boas estimativas dos algoritmos propostos, para a base de dados analisadas e na presença de muitos modelos sem sofrerem grandes interferências da escolha dos limiares.

Severe Acute Respiratory Syndrome 2 (SARS-CoV-2) comprises one of the complications triggered by the new coronavirus. This study aims to propose a comparison between two machine learning-based models in different contexts to predict mortality in cases of Severe Acute Respiratory Syndrome (SARS) caused by the 2019 coronavirus, COVID-19. The data used are available on the DataSUS platform and cover the period from January 2021 to December 2022. Consequently, a descriptive statistical analysis, variable selection, and the development of two models were carried out, one before the milestone of the second dose of COVID-19 vaccination and another after. Regarding the metrics, model 1 presented an accuracy of 71.8%, while model 2 achieved an accuracy of 80%, thus contributing to the decision-making process for tackling the disease.