Raw Material Testing Validation & Verification for SQF

Hi Amigos!

I did it!  I got a list of all of our raw materials, and the vendors!  And the specs!  And the COAs!  AND my receiving department isn't accepting things without a COA!  Hooray! 

So now I'm at the part where I take the COA and do raw material testing to check it to see if it's right.  I don't have a micro lab and micro testing on all incoming ingredients is ridiculously impractical - BUT

I have three cubicle walls, and I can totally, easily, quickly, and cheaply test a lot of common COA specs - pH, color (if analyzer spec), brix, viscosity, moisture, fat content if I really wanna blow the budget on fancy equipment. 

I would like to test these specs for incoming ingredients and reject  the ingredient if it doesn't meet spec.  This is practice I have experience with and I have seen inaccurate COAs, which means I know there's a reason and we gotta make sure they aren't ripping us off and selling us something 20% extra water.  Or 5% thinner.  Or acrid with a low pH from rotting. It happens!  Etc.

It would not be practical to send these out to a 3rd party lab (for pH?  Really?).  I have my instruments on a regular calibration schedule with certificates, logs of all "satellite lab" activities, white papers,and  receiving and lot tracking info for all testing materials.  I calibrate my pH meter every time I use it. If something is out of spec, I resample and retest. I have SOPs for raw material testing. I was a microbiologist for years so my stuff comes correct.  I use AOAC standard methods.

Is it necessary, to ensure food safety and quality, comply with the standards set forth in the most current SQF code AND general best manufacturing practices, to further validate or verify the methodology or equipment I use, or is the current documentation sufficient?

If so, what kind of record or study would that be?  I mean, I have them calibrated, I have the white papers.  Is it necessary to send a duplicate set samples to a 3rd party lab to ensure they are reproducing results or obtain a laboratory certification?  I'm really lost as to what section of the code to reference as I'm not used to SQF standards and they seem vague and open to wild interpretation to me.

magenta,

I'm curious... with all the documentation you have from your suppliers, why do you feel the need to conduct further testing?

My understanding of the SQF Code is you do either COA's or in-house testing, not both. Seems like you will be opening yourself up to having to meet the in-house lab standards as well.

Oh BTW, congratulations on getting all that documentation and on training the receiving folks to not receive without COA's. I know how tough that can be. ;-)

Phil

Dear magentam,

I admire yr enthusiasm.

But as per previous post, I think most people seek COAs,etc to lessen their workload. Probably including SQF auditors.

Nonetheless some Companies do analyse everything coming in (within technical/financial credibility). Particularly the ones who have a lot to lose in the event of a problem and are already "not poor". In other words it's a risk-based assessment. Some Companies analyse everything going out also, every day of production.  So much for HACCP.

We live in a pragmatic Universe.

Rgds / Charles.C

Hi all,

I don't accept any raw materials until I receive a SQA / COA. I have an internal lab which does the following test: Moisture content , Protein Kjeldahl, Fats content by hydrolysis, Ash content , calcium content , phosphorus content, rapid afflatoxins test, microscopic tests which are done on all raw materials (where applicable). Due to budget contraints , we cross check the COA with external labs to ensure that our suppliers are meeting our expectations. We came across a lot of results (tested by our extenal accrediated labs) which  were different from COA of our suppliers.

In these cases, we had to revert back to suppliers.

All our tests methods are according to AOAC methods.

Rgds,

Rudra

Hiya

I'm with you on this Mags

I have seen my fair share of CoA's which are basically plucked from the air, and I do the same as you, I test, albeit stuff like colour, salt, moisture fats, etc inhouse. I even send out samples to our external lab for micro testing. We hold the product here until its cleared.

With respect to your in house testing, I would carry out a proficiency test every 6 months or so. Get a sample, get everybody who tests to test it, and send a portion out to an external lab to test. Then collate al the results, work out the standard deviations and then the Z Scores.  Simples way to validate.

Drop me a mail if you need any other info (like how to work out the Z Scores!)

Caz x

Thanks:-)

Would really appreciate to know more about the Z score..

Rudra

My understanding of the SQF Code is you do either COA's or in-house testing, not both. Seems like you will be opening yourself up to having to meet the in-house lab standards as well.

Yes , one major issue going forward with SQF is that my management doesn't want to start programs that are not strictly required because then we will be audited against them, or then we will have to comply with the standards :/ .  This is also their argument against going to level 3 vs level 2 and implementing a variety of programs.  Interestingly, management seems evenly divided between people who only want to meet minimum requirements and people who want to go for quality / food safety culture, and the floor & owner want to go level 3. Not only do I want to open my self up to the in house testing requirements, I want to do what it takes to fully meet them. 

I'm curious... with all the documentation you have from your suppliers, why do you feel the need to conduct further testing?

Phil

I like to check COAs because raw materials are very expensive and I want to make sure I'm getting what I pay for.   As we all know from our customer complaints, products that aren't perfect sometimes get released.   I'd like to know about those problems before we use it in product.

Mistakes happen!  If a vendor corrects  it, hooray, if not...I know not to buy from them.   COAs are wrong a lot. 

Oh BTW, congratulations on getting all that documentation and on training the receiving folks to not receive without COA's. I know how tough that can be. ;-)

Phil

Thanks!  You should have seen their faces when they couldn't refrigerate the butter upon arrival due to no COA, another rule I've beat into them.  Talk about a moral crisis XD.

Dear magentam,

 

I admire yr enthusiasm.

 

But as per previous post, I think most people seek COAs,etc to lessen their workload. Probably including SQF auditors.

 

Nonetheless some Companies do analyse everything coming in (within technical/financial credibility). Particularly the ones who have a lot to lose in the event of a problem and are already "not poor". In other words it's a risk-based assessment. Some Companies analyse everything going out also, every day of production.  So much for HACCP.

 

We live in a pragmatic Universe.

 

Rgds / Charles.C

Ah, very interesting and on the nose argument against testing.  Yes, it's true that suppliers make COAs so that I don't have to.  And these specs aren't critical to my processing.  If the workload becomes overwhelming, I'll look into reducing testing by risk categories.

Hiya

 

I'm with you on this Mags

 

I have seen my fair share of CoA's which are basically plucked from the air, and I do the same as you, I test, albeit stuff like colour, salt, moisture fats, etc inhouse. I even send out samples to our external lab for micro testing. We hold the product here until its cleared.

 

With respect to your in house testing, I would carry out a proficiency test every 6 months or so. Get a sample, get everybody who tests to test it, and send a portion out to an external lab to test. Then collate al the results, work out the standard deviations and then the Z Scores.  Simples way to validate.

 

Drop me a mail if you need any other info (like how to work out the Z Scores!)

 

Caz x

Yes!  Proficiencies!  That's the one I forgot / the name of sending duplicate samples to the lab! Thank you!  There are a number of companies who sell proficiency testing with certificates as well if I really wanted to go all out.

It seems like statistical analysis & z-scores is a skill a few others want to learn as well

Ok

this is the kind of thing that I do

I use an online program to calculate the Z Score http://www.danielsop...calc.aspx?id=22

Your population mean is the result the operator had, the Standard deviation  is just that, and the value is the external lab result

Voila!

Z Scores should be between +2 and -2

I do a nice pretty graph of it all then!

I'm sure you can do a google search as to why the Z Score should be between +2 & -2; can't spoon feed you all the info!

Caz x