Carotenoids are pigments that may be used as colorants and antioxidants in food, pharmaceutical, and cosmetic industries. Since they also benefit human health, great efforts have been undertaken to search for natural sources of carotenoids, including microbial ones. The optimization of culture conditions to increase carotenoid yield is one of the strategies used to minimize the high cost of carotenoid production by microorganisms. Halophilic archaea are capable of producing carotenoids according to culture conditions. Their main carotenoid is bacterioruberin with 50 carbon atoms. In fact, the carotenoid has important biological functions since it acts as cell membrane reinforcement and it protects the microorganism against DNA damaging agents. Moreover, carotenoid extracts from halophilic archaea have shown high antioxidant capacity. Therefore, current review summarizes the effect of different culture conditions such as salt and carbon source concentrations in the medium, light incidence, and oxygen tension on carotenoid production by halophilic archaea and the strategies such as optimization methodology and two-stage cultivation already used to increase the carotenoid yield of these microorganisms.
A Transformação Digital (TD), também chamada de Indústria 4.0, tem implementado tecnologias para o aperfeiçoamento dos processos produtivos e de gestão organizacional. Diante desse cenário, o presente estudo pretende verificar como a Gestão do Conhecimento (GC) pode contribuir para a TD nas organizações e, ao mesmo tempo, como a TD está colaborando para a GC. Sendo assim, primeiramente a GC é contextualizada junto aos conceitos de TD. Então, para se atingir o objetivo deste estudo, uma revisão sistemática da literatura foi realizada, considerando a relação dos pilares da Indústria 4.0 e dos Sistemas Ciber-Físicos com a GC. Para finalizar, os resultados da revisão sistemática são analisados e discutidos para responder ao objetivo deste trabalho.
2012) A computational control implementation environment for automated manufacturing systemsThis paper presents a computational environment as a tool for supporting the implementation of control coding of an automated manufacturing system. The proposed environment considers a cyclic three-stage control development -modelling, synthesis and implementation -until the real system accomplishes the required specification, resulting in the automated and integrated manufacturing system. The research details the three stages and describes the steps executed for each one. The mathematical formalism used in this work is also presented, as a basis for control implementation. The implementation environment is proposed in order to validate the control structure of this formalism and to allow a progressive integration of control hardware and software. To submit to a test and validate the proposal environment, two experiments are performed, in two different manufacturing systems. Thus, it is demanded that the control system can be reconfigurable in a fast and reliable way.
Predictive maintenance can be used for the optimization of machine availability, reducing maintenance costs, improving quality management and decision-making. This paper proposes the use of 5 time-series forecasting methods, being two classical approaches: Exponential Smoothing and Autoregressive Integrated Moving Average (ARIMA); and three based on machine learning: Radial Basis Function Neural Networks (RBF-NN), Support Vector Regression (SVR), and Long-Short Term Memory (LSTM). The meta-parameters of the models were optimized using Differential Evolution (DE) algorithm. In the presented analysis, a quality inspection database from an automaker factory floor was presented. The main idea was to predict deviations from tolerance, anticipating corrective actions in the production process. As a result, it was possible to compare the efficiency of the 5 models in relation to three time series analyzing their characteristics, strengths and drawbacks. Resumo: A manutenção preditiva pode ser usada na otimização da disponibilidade de máquinas, redução de custos de manutenção, gestão da qualidade e processo de tomada de decisões. Neste trabalho realizouse o estudo de 5 modelos de previsão de séries temporais, dois clássicos: Suavização Exponencial e Autorregressivo Integrado de Médias Móveis (ARIMA), e três baseados em aprendizado de máquina: Rede Neural de Base Radial (RBF-NN), Regressão por Vetores de Suporte (SVR) e Memória de Curto e Longo Prazo (LSTM). Os meta-parâmetros dos modelos foram otimizados usando o algoritmo de Evolução Diferencial (DE). A base de dados utilizada neste estudo é proveniente da inspeção da qualidade no chão de fábrica de uma montadora. Neste contexto, a ideia principal foi prever desvios relacionados à tolerância, antecipando ações de correção no processo produtivo. Como resultado, foi possível comparar a eficiência dos 5 modelos em relação a três séries temporais analisando suas características, vantagens e desvantagens.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.