| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 947.56 KB | Adobe PDF |
Abstract(s)
Este relatório, decorrente de um estágio, aborda aspetos na validação de
métodos internos para análises de proteínas e as funções desempenhadas no controlo
produtivo, por um técnico de laboratório na Schreiber Foods, uma empresa laticínios.
O objetivo principal foi avaliar a viabilidade do uso do método Dumas, de forma a
substituir o método Kjedahl no contexto da empresa, visando uma maior rapidez,
segurança e sustentabilidade ambiental. O estudo abrangeu três tipos de produtos: o
leite, o semielaborado e o iogurte. Estes representam as três fases do processo de
fabrico de iogurte. Foram utilizadas cinco fontes de análise, uma para o método Dumas,
VELP NDA 701, e quatro para o método Kjedahl, “Lab1” e “Lab2” que representam
empresas externas acreditadas, e MS120 e FT3 equipamentos internos de análise
calibrados. As análises efetuadas sugerem uma forte correlação entre os métodos,
especialmente para o semielaborado, 0,994, e iogurte, 0,988. Com recurso a análise de
regressão foi possível estabelecer fatores de conversão entre os métodos de 0,1608 para
o semielaborado e 0,0578 para o iogurte, embora com limitações fora do espaço
amostral. Adicionalmente, testes ANOVA confirmaram diferenças significativas entre
os métodos, bem como a sua independência face a outros fatores. No entanto, o método
Dumas apresentou uma baixa precisão com erro médio de ±0,11 fora dos limites
exigidos pela empresa Schreiber Foods para a sua aplicação a nível empresarial.
This report, resulting from an internship, addresses aspects in the validation of in-house methods for protein analysis and the roles played in production control by a laboratory technician at Schreiber Foods, a dairy company. The main objective was to assess the viability of using the Dumas method to replace the Kjeldahl method within the company, aiming for greater speed, safety, and environmental sustainability. The study covered three types of products: milk, semi-processed products, and yogurt, representing the three phases of the yogurt manufacturing process. Five sources of analysis were used, one for the Dumas method, VELP NDA 701, and four for the Kjeldahl method,"Lab1" and "Lab2," representing accredited external companies, and MS120 and FT3, internally calibrated analysis equipment. The conducted analyses suggest a strong correlation between the methods, especially for “semielaborated” products, 0,994, and yogurt, 0,988. Regression analysis allowed for establishing a conversion factor between the methods for “semielaborated” products being 0,1608 and for yogurt, 0,0578, although with limitations outside the sample space. Additionally, ANOVA tests confirmed significant differences between the methods, as well as their independence from other factors. However, the Dumas method exhibited low precision average error of, ±0.11 outside the scope required by the Schreiber Foods Company for its application at an enterprise level.
This report, resulting from an internship, addresses aspects in the validation of in-house methods for protein analysis and the roles played in production control by a laboratory technician at Schreiber Foods, a dairy company. The main objective was to assess the viability of using the Dumas method to replace the Kjeldahl method within the company, aiming for greater speed, safety, and environmental sustainability. The study covered three types of products: milk, semi-processed products, and yogurt, representing the three phases of the yogurt manufacturing process. Five sources of analysis were used, one for the Dumas method, VELP NDA 701, and four for the Kjeldahl method,"Lab1" and "Lab2," representing accredited external companies, and MS120 and FT3, internally calibrated analysis equipment. The conducted analyses suggest a strong correlation between the methods, especially for “semielaborated” products, 0,994, and yogurt, 0,988. Regression analysis allowed for establishing a conversion factor between the methods for “semielaborated” products being 0,1608 and for yogurt, 0,0578, although with limitations outside the sample space. Additionally, ANOVA tests confirmed significant differences between the methods, as well as their independence from other factors. However, the Dumas method exhibited low precision average error of, ±0.11 outside the scope required by the Schreiber Foods Company for its application at an enterprise level.
Description
Keywords
Análise Proteica Iogurte Método Dumas Método Kjedahl