Lab Report
Product: Swiss Cheese
Analyzer: ZX-880 Near-Infrared Food/Cheese Analyzer

INTRODUCTION

Approximately 30 Swiss Cheese samples were forwarded to Zeltex on 9/28/00 from Hans Rothenbuhler, Inc. These samples were sent initially without analysis. The analysis followed approximately three weeks later.

These samples were run on a ZX880 and the following results are obtained:

PROTOCOL

The ZX880 is designed to be used with any of four different cups. For cheese, we used the clamshell cup with zip-lock bags for easy handling and clean-up. Following the instructions from the company, we pulverized the samples in a blender to obtain a uniform consistency (Figure 1), spooning the sample into a zip-lock bag, (Figure 2), placing zip-lock bag in the clamshell cup ( Figure 3)and place the clamshell cup with instrument (Figure 4).

Figure 1	Figure 2
Figure 3	Figure 4

The optical data (Figure 5) for these samples was then saved to a computer using the Zeltex software. After collecting the optical data for all the samples, this optical data was then merged with the lab analysis and reduced by using Multiple Linear Regression (MLR). Once the MLR was completed, the following reductions were developed (Figures 6, 7, 8 and 9).

Figure 5 - Optical Data

Figure 6 - Oil	Figure 7 - Protein
Figure 8 - Moisture	Figure 9 - Salt

As you see from these reductions, the Standard Error for protein, salt and Oil is within the acceptable range. The moisture Standard Error was a little high. Since moisture is the easiest constituent to analyze in near-infrared, it is believed the poor accuracy is due to the amount of time Zeltex stored these samples in the refrigerator which allowed moisture to seep out of the sample. This seeping affected the protein and oil, increasing the Standard Deviation slightly.

CONCLUSIONS

The ZX880 will work extremely well after the cheese has been sent through a blender. This process is necessary since the Swiss cheese cannot be cored and placed within the cup due to the holes in the product.

Since Swiss cheese, when first formed, has no holes, this product could be done using a cutter to get a standard-size sample and placed within the clamshell cup. Blenderizing the sample would not be necessary, but this would require a separate calibration for cheese with holes and without holes. You should be able to do a single calibration on Swiss cheese if the blender is used for sample preparation on all samples.