NIR Analysis of Cheese—A Fonterra Perspective

Determination of fat, protein, moisture, and salt content of Cheddar cheese using mid-infrared transmittance spectroscopy

“…Although using NIR for cheese composition analysis is 925 appropriate in terms of the ease of sample preparation and speed of analysis, the work required to develop calibrations for each cheese type within each cheese factory is not practical or cost effective. Holroyd (2011) indicated that there has been an evolution of equipment manufacturer-specific NIR calibration approaches using multiple linear regression and PLS over the years. If a factory has more than one NIR analyzer within their laboratory, separate calibration development may be required for each instrument.…”

“…The accuracy of each prediction model for each cheese type is dependent on the accuracy of the reference chemistry performed in the cheese factory and the concentration range and distribution of fat, protein, and moisture in the specific calibration samples used for PLS model development (McKenna, 2001; Barbano and Lynch, 2006). Consistency of sample presentation (particle size distribution) is often a source of analytical variation in NIR cheese analysis (Holroyd, 2011). In a review of NIR analysis of cheese, Holroyd (2011) gave the Fonterra Research Centre (Palmerston North, New Zealand) perspective on the evolution of NIR analysis of cheese for a large group of cheese factories in New Zealand over the last 30 yr, and on the evolution of NIR analytical technology.…”

“…In a review of NIR analysis of cheese, Holroyd (2011) gave the Fonterra Research Centre (Palmerston North, New Zealand) perspective on the evolution of NIR analysis of cheese for a large group of cheese factories in New Zealand over the last 30 yr, and on the evolution of NIR analytical technology. Holroyd (2011) reported that the following factors are important in NIR cheese analysis: 1careful sample preparation using several cheese cores from each block, (2) representative sampling of the block, (3) a knowledge of the sample matrix, and (4) sample preparation that provides a uniform presentation of the sample to the NIR. The accuracy of analytical performance of an indirect measurement method's prediction of reference chemistry is called the standard error of prediction (SEP).…”

“…The accuracy of analytical performance of an indirect measurement method's prediction of reference chemistry is called the standard error of prediction (SEP). Holroyd (2011) reported the SEP metrics from many studies of NIR cheese analysis performance ranging from 0.2 to 0.5% for fat, protein, and moisture during the period from 1993 to 2003. In 2003, Fourier transform NIR instruments became more common and some individual companies developed global PLS models for cheese analysis that reduced the amount of reference chemistry, time, and cost required at the local level for calibration, while maintaining SEP values more consistently in the range of 0.2 to 0.35% for fat and moisture prediction (Holroyd, 2011).…”

Fast and green method for analysis of Serra da Estrella

“…The reflected rays contain the spectral information of the sample (Burns, Ciurczak, 2008). It is well established that fat, moisture and protein composition of cheese can be measured successfully using NIR spectroscopy (Holroyd, 2011). The classical methods used in the determinations of fat and protein in food have numerous disadvantages, like being time and energy consuming, having limited sample throughput and leading to possible harmful effects on the environment, because of the large amount of chemicals required.…”