CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Determinación de humedad y actividad de agua de un producto tipo salami (fuet, salchichón) mediante un equipo NIR low cost

Estudiant que ha llegit aquest projecte:

Tutor/a o Cotutor/a: HERNANDEZ YAÑEZ, EDUARD

Departament: DEAB

Títol: Determinación de humedad y actividad de agua de un producto tipo salami (fuet, salchichón) mediante un equipo NIR low cost

Data inici oferta: 01-02-2020      Data finalització oferta: 01-10-2020

Estudis d'assignació del projecte:
    MU MKET4FOOD+BIO

Lloc de realització:


En empresa (cal signar un conveni de cooperació)

        Tutor/a Extern: Josep Comaposada
        Institució/Empresa: IRTA

Paraules clau:
NIR, producto cárnico, low cost

Descripció del contingut i pla d'activitats:
Proyecto de determinación de humedad y actividad de agua de un
produto tipo salami (fuet, salchichón) mediante un equipo NIR
low cost. En este proyecto se debe de muestrear el producto en
una empresa cárnica y luego realizar los análisis en el
laboratorio. Una vez obtenidos los resultados analíticos de
humedad y actividad de agua, se debe de crear el modelo de
calibración del equipo NIR mediante un software informático.
También se debe de crear un modelo matemático que permita
relacionar los parámetros analizados en distintos puntos

Overview (resum en anglès):
The agri-food sector is one of the main economic engines in Europe. This sector includes the meat industry such as the production of cured sausages. The meat industry is committed to the development of new products and the establishment of quality and food safety controls to transfer to consumers the benefits and added value of its products. Therefore, it is important to implement innovative integrated production techniques and continuous improvement throughout the entire
food chain. Quality controls at all levels of the production process are also important. Near infrared spectroscopy (NIRS) has demonstrated its ability to analyze food without altering its properties or destroying the analyzed piece or part of it. In addition, they are considered clean technologies, since they do not generate waste.
The main objective of this work was to determine the potential of near infrared spectroscopy (NIRS) technique to be an adequate tool for predicting water activity (aw) and humidity in a cured meat product such as fuet in different parts: on the surface of the fuet, in the center of the fuet or in the minced fuet. To achieve this objective, the work was developed in different stages. Initially,the NIR spectra of the fuets were taken in different parts in a sausage factory using the SCiO
pocket NIR instrument, then the aw and moisture values of the samples were determined and finally, with both kind of information, the prediction models were created and evaluated. The spectral and composition measurements were evaluated in the three variants (surface, center, and
minced). And predictive models were then developed to determine moisture and water activity, by using the multivariate regression method PLS. Also, several data preprocessing protocols were applied in the spectra as a try to obtain the best calibration and prediction models with the minimum prediction error. The effect of the packaging film on the results of the prediction models
was also studied. Furthermore, the performance of the NIR SCiO spectrophotometer was compared with that of an- house NIR Hamamatsu sensor, based on the values of coefficient of determination and the prediction errors of both devices.
The obtained results showed that the NIR SCiO sensor provides a different response depending on the area analyzed (surface, center or minced), which is attributed to the change of matrix, the variation of the composition (salt / humidity, fat) and the structure of the sample, involving a different light scattering. At the same time, differences in precision have been observed due to the
chosen spectral pretreatments. The prediction errors of the water and moisture activity obtained in
the models of the three variants were considered adequate, where the coefficients of determination
of prediction R2p were is above 0.97 for all preprocesses. Also, it has been shown that it is feasible
to predict the aw and humidity of the minced samples even with the presence of the packaging film. The cross validation errors of the prediction models developed for the determination of water activity were similar, between 0.0039 with film and 0.0036 without film. But, they were different for the prediction of humidity, between 1.24 % y 1.71 % for the samples with film and without
film respectively. Finally, the best prediction models of water and moisture activity in the case of samples with film was obtained with the NIR Scio spectrometer rather than the Hamamatsu spectrometer. The prediction coefficient with Scio was 0.9932 and 0.9925 for water activity and humidity respectively. With Hamamatsu sensor, the prediction coefficients were a slightly lower
has decreased for both parameters of (0.9658 and 0.982, respectively).In brief, the results created revelaed that from the NIRS technology have proven its feasibility in
accurate estimation of various food quality parameters and could represent an improvement in the control of drying systems of meat industry.

© CBLTIC Campus del Baix Llobregat - UPC