Homogeneidade da mistura na fabricação de premix para alimentação animal: efeito do volume de ocupação e tipo de misturador

• Ano de defesa: 2024
• Autor(a) principal: Sousa, Jonas Rogério de Melo
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia de Alimentos
• Palavras-chave em Português: Nutrição animal; homogeneidade da mistura; coeficiente de variação (CV); mistura de pós; misturadores horizontais; Animal nutrition; coefficient of variation (CV); mixture homogeneity; powder blending; horizontal mixers; CNPQ::CIENCIAS AGRARIAS::CIENCIA E TECNOLOGIA DE ALIMENTOS; Tecnologia de alimentos; Alimentos de origem animal; Alimentos - Biotecnologia; ODS::ODS 3. Saúde e bem-estar - Assegurar uma vida saudável e promover o bem-estar para todos, em todas as idades.
• Link de acesso: https://repositorio.ufu.br/handle/123456789/44791; http://doi.org/10.14393/ufu.di.2025.5503

Animal production has reached significant levels of productivity, much of which is due to nutrition, which represents more than 70% of production costs. Therefore, the manufacturing process of animal feed products requires high levels of control, especially regarding the quality of the mixtures, ensuring that ingredients are adequately dispersed within the batch of feed produced to ensure homogeneous nutrient consumption by the animals. Operational characteristics, such as the mixer’s fill level and the type of mixer, can affect the homogeneity of the mixture. Additionally, depending on the formulation, the mixing process can be more or less challenging. Thus, this study aimed to evaluate the impact of the fill level in two types of horizontal mixers—a paddle mixer and a ribbon mixer—on the homogeneity of a vitamin-mineral premix. The fill levels of 10%, 30%, 50%, 70%, and 90% were evaluated in relation to the total usable volume of each mixer. The coefficient of variation (CV) was used as a method to assess mixture homogeneity, using calcium, copper, iron, magnesium, manganese, sodium, sulfur, selenium, and zinc as indicators, along with a composite indicator based on the average of all the mineral elements analyzed. The decision criterion was to obtain a CV below 10%. All tested fill levels in both mixers met the decision criterion, except for sodium in the ribbon mixer at 10% fill level, where the CV was 10.6%. No differences were observed between the tested fill levels in the paddle mixer by regression analysis (P>0.05). In the ribbon mixer, a quadratic effect was observed (P<0.05) for copper, iron, manganese, sulfur, and the composite indicator, and a linear effect (P<0.05) for sodium, selenium, and zinc, with the highest CV results obtained at 10% fill level for all tested indicators. When comparing the results between mixers, an interaction was observed between mixer type and fill level for manganese and selenium (P<0.05). For calcium, copper, and iron, the CV was lower (P<0.05) in the ribbon mixer. Regardless of the mixer type, the fill level had a quadratic effect (P<0.05) on zinc and a linear effect (P<0.05) on iron, sodium, and the composite indicator. When the data were analyzed using machine learning, the decision tree revealed that the best CV results (below the median of 5.425%) for the composite indicator are obtained when the premix is manufactured in the ribbon mixer with a fill level greater than 20%, with 90% accuracy.