3 replies to this topic

[Montana Milling Inc]

Hello all.  Hope this finds all of you safe and well.

 

Questions regarding 5.3.8 of the BRC Standard for Food Safety and allergens.  We are a flour manufacturer and do not purchase or process soy.  However, one of our customers has requested us to mill a product that contains sprouted soy.  This would account to less than .5% of total product through this system annually.  We purge product though this system to clean it between production runs.  

 

(1) Can I conduct a swab test annually to meet the "validation" requirements of 5.3.8 to substantiate the line does not have soy present.

 

(2) Then conduct a strip test after runs of this product, positive at 2 ppm, to substantiate soy is not present after a purge is conducted.

 

Thanks for your help and insight.

[Tony-C]

Hi Benjamin,

 

For validation, I would also sample the first product manufactured in the next production run.  

Validation data will need to be collected from several production runs to ensure it is representative.

Trials should be worst-case production/purging trials, where you swab targeted test locations including food contact surfaces and difficult-to-clean areas.

If you use the strip test as well during your validation, you can correlate the two and show that it is an acceptable method for verification.

 

Kind regards,

 

Tony

[Montana Milling Inc]

Thanks Tony!  I appreciate the help

[VThiruselvi]

Hi Benjamin,

 

For validation, I would also sample the first product manufactured in the next production run.  

Validation data will need to be collected from several production runs to ensure it is representative.

Trials should be worst-case production/purging trials, where you swab targeted test locations including food contact surfaces and difficult-to-clean areas.

If you use the strip test as well during your validation, you can correlate the two and show that it is an acceptable method for verification.

 

Kind regards,

 

Tony

Hi Benjamin,

 

As Tony mentioned here, it should be contact and product for few runs to validate there is no cross contamination.and verification in specific intervals