20 replies to this topic

[margreteva]

Hello, I work for a meat processing plant in Iceland. We are currently working towards getting BRC certified. Our biggest issue at the moment I would say is the metal detection. We have a metal detector but we have so many products of different sizes and weights (some boneless and some not) running through every minute that the metal detector doesn't work for us.

Does anybody here have any experience with being BRC certified, but not using a metal detector (or x-ray) for their products. 

I'm thinking if it would be possible for us to omit the metal detection step if we establish a strict knife policy and also start with inspections of equipment which could be a source of metal contamination. Additionally, we would use the metal detector for our minced meat, goulash and such products whereas it would be difficult for the consumer to spot a metal contamination by eye.

We have not had an incident with metal contamination for the last year (one incident a year ago where a metal was found on a piece of meat which had not been packed yet) and have not received a complaint thereof at least for the last few years. 

Thank you for your comments! :) 

[QAGB]

I hope you get some good responses from BRC certified folks that don't use a metal detector. Speaking as one that has dealt with BRC and had metal detectors in place, I would not advise not having one. I would think you have a lot of potential for metal contamination with your cutting utensils and processing equipment. A strict knife policy really isn't that easy to implement. It can be done, but there's always going to be a problem. What happens when your employee returns a knife missing a chunk of its blade? 

 

Admittedly, I am not from a meat background, so I am going to ask this question. Why do you have multiple variances of products on the same line at the same time? I would have thought you'd be running "like" products. If you're packing a production run of boneless meat, wouldn't you want all your products to be boneless? Similarly, if you are packing 2 lb. sizes, wouldn't you want all your products to be cut to ~2 lbs.?

 

Metal detectors can be setup per product type. If you were running like products, you could setup your metal detector per product. 

[SQFconsultant]

I did an inspection on a facility that took their metal detector out when they came up eith whst they thought was a great sharps program until they had an issue one day with 2 pieces of boning knives missing... that were unfortunately found by two customers - one piece found prior to consumption with the other in the person's mouth after causing extensive dental and gum damage. The company put in a customized metal detection system that automatically scans each piece of product first and then automatically adjusts the MD to handle that piece. Expensive solution, but much better than causing damage, choking or killing someone.

[margreteva]

Thank you for your reply!

 

The reason for this problem is that we don't really have product lines. We have a meat cutting room where the carcasses come in. There are several people working in the room cutting the meat and each one is working on a different part of the carcass. Each person then puts their product on the same conveyor belt which takes it to the packing room. There we have people putting the different products in different vacuum bags or in some cases, large laminated carton boxes. The products in the vacuum bags then go to the vacuum machine but after that they go on the same conveyor belt that takes every product past the scales, through the metal detector and then into the cold storage room. 

We have tried programming the metal detector with various products, but it doesn't work because the products vary so much.

[margreteva]

I did an inspection on a facility that took their metal detector out when they came up eith whst they thought was a great sharps program until they had an issue one day with 2 pieces of boning knives missing... that were unfortunately found by two customers - one piece found prior to consumption with the other in the person's mouth after causing extensive dental and gum damage. The company put in a customized metal detection system that automatically scans each piece of product first and then automatically adjusts the MD to handle that piece. Expensive solution, but much better than causing damage, choking or killing someone.

 

Thank you for your reply! This system is something I will definitely look into

[QAGB]

Thank you for your reply!

 

The reason for this problem is that we don't really have product lines. We have a meat cutting room where the carcasses come in. There are several people working in the room cutting the meat and each one is working on a different part of the carcass. Each person then puts their product on the same conveyor belt which takes it to the packing room. There we have people putting the different products in different vacuum bags or in some cases, large laminated carton boxes. The products in the vacuum bags then go to the vacuum machine but after that they go on the same conveyor belt that takes every product past the scales, through the metal detector and then into the cold storage room. 

We have tried programming the metal detector with various products, but it doesn't work because the products vary so much.

 

 

I understand now. That does make for a predicament. You really should look into SQFconsultant's advice for the metal detector. I would not advise against not having metal detection, because I think your risk is certainly high enough to warrant one. It'll be costly to put in a system like that, but sure does beat the alternatives.

[Charles.C]

Metal detection equipment shall be in place unless risk assessment demonstrates that this does not improve the protection of final products from metal contamination.

Where metal detectors are not used justification shall be documented.

The absence of metal detection would only normally be based on the use of an alternative, more effective method of protection (e.g. use of X-ray, fine sieves or filtration of products).

 

Hi margreteva,

 

BRC are well-known to be highly "sensitive" regarding Foreign materials, ie Clause 4.10.

 

Presumably to emphasize their focus, BRC previously issued a metal detection decision tree (2012), see this Post -

 

https://www.ifsqn.co...ed/#entry114807

(I have no idea if subsequently updated ?)

 

The current text in 4.10.3.1 seems to triply illustrate variations of the above emphasis.

 

 

So what do the BRC, I.G. Guidelines say about options to use/not use a MD ?

 

Is it feasible to use a different, GFSI-recognized, FS Standard ?

[zanorias]

The I.G for 4.10.3.1 states as below. It also includes the decision tree for metal detection which appears unchanged from the version linked in Charles's above post.

 

Requirement for metal detection

The Standard presumes that metal detection provides improved protection for customers and should form

part of the food protection system of a site. Its absence would normally only be based on the use of an

alternative, more effective, method of protection (e.g. the use of X-ray, fine sieves or filtration). There will,

however, be situations where metal detection does not, on the basis of risk assessment, provide any

significant additional protection to the consumer.

Where metal detectors are not used, a risk assessment must be available to justify the reasons why. While

complaint levels are a factor in making a decision on the necessity for a metal detector, this evidence alone

will not be sufficient justification for not using one. (For example, there may be instances of contamination

which have not been reported by consumers.) Any justification for the absence of metal detection should be

based on the nature of the product, the risk to the consumer, and alternative controls in place at the site

which prevent metal contamination. Cost alone is not sufficient reason.

 

 

I'm thinking if it would be possible for us to omit the metal detection step if we establish a strict knife policy and also start with inspections of equipment which could be a source of metal contamination. Additionally, we would use the metal detector for our minced meat, goulash and such products whereas it would be difficult for the consumer to spot a metal contamination by eye.

We have not had an incident with metal contamination for the last year (one incident a year ago where a metal was found on a piece of meat which had not been packed yet) and have not received a complaint thereof at least for the last few years. 

 

My concern with the above mentioned controls are the susceptibility to human error; it relies on the knife policy being followed correctly and inspections conducted by human eye. What about the raw material already being contaminated with metal? The key question here, reviewing the BRC guidelines, is: Do these controls establish a more effective method of protection than a metal detector? Until you can answer 'yes' to that I struggle to see how it will fly with BRC.

[pHruit]

We're a BRC site that doesn't use a metal detector, but we have a very different product range to yours. Every year the auditor asks us to provide risk-based justification for this (even if it's the same auditor as the previous year, looking at their notes from last year) so they really are very focussed on this. We did investigate MD options and repeat this every few years, so our assessment is based on the quoted performance specs provided for these vs. our current sieving/filtration systems - the latter have a smaller "detection" limit and this is therefore accepted by BRC.

Given your product types I cannot see how you'd justify the position of not using metal detectors - "it's too complex" is unlikely to be an acceptable basis.

[GMO]

Sooo tricky... strictly speaking, yes you can take it out but you will be fighting in every audit.

 

Maybe can you talk us through what you've done to get the metal detector operational and perhaps we can help?  A test piece size up to 7mm is defensible. 

 

I can see why x-ray would be a pain in the bottom for you though with raw bone.  I'm sure x-ray methodology has moved on a long way in the time since I last used any (probably 12 years ago?)  It could also have the benefit of being able to test for bone in boneless cuts?

 

Then there is the risk to consider that if people know you don't have metal detection and they're annoyed with the company....   How would you defend that decision in court. 

 

Tricky... Doable in theory but honestly I wouldn't.  I'd find a way to make metal detection or x-raying work.

[GMO]

We're a BRC site that doesn't use a metal detector, but we have a very different product range to yours. Every year the auditor asks us to provide risk-based justification for this (even if it's the same auditor as the previous year, looking at their notes from last year) so they really are very focussed on this. We did investigate MD options and repeat this every few years, so our assessment is based on the quoted performance specs provided for these vs. our current sieving/filtration systems - the latter have a smaller "detection" limit and this is therefore accepted by BRC.

Given your product types I cannot see how you'd justify the position of not using metal detectors - "it's too complex" is unlikely to be an acceptable basis.

 

Yes, I think your example is justifiable because of the filtration to a tighter limit.  I can't see that justification here though.

[pHruit]

Yes, I think your example is justifiable because of the filtration to a tighter limit.  I can't see that justification here though.

Indeed - but even then the justification is regularly challenged. It's certainly not a position I'd want to try to defend from the OP's shoes!

[Kiran]

Hi Margreteva,

 

there is mention of knife policy & controls, and some have already commented on this point.  Is there risk of metal pieces such as lead shots, broken needles in carcass?

 

x-ray is a possible option as well.  It can have dynamic product switch (change product set up depending on which SKU is entering x-ray machine) and would be able to detect bones in boneless products as well.

[margreteva]

Sooo tricky... strictly speaking, yes you can take it out but you will be fighting in every audit.

 

Maybe can you talk us through what you've done to get the metal detector operational and perhaps we can help?  A test piece size up to 7mm is defensible. 

 

I can see why x-ray would be a pain in the bottom for you though with raw bone.  I'm sure x-ray methodology has moved on a long way in the time since I last used any (probably 12 years ago?)  It could also have the benefit of being able to test for bone in boneless cuts?

 

Then there is the risk to consider that if people know you don't have metal detection and they're annoyed with the company....   How would you defend that decision in court. 

 

Tricky... Doable in theory but honestly I wouldn't.  I'd find a way to make metal detection or x-raying work.

 

Last year we got a guy from a company here in Iceland which specialises in validating and verifying metal detectors. He set up a few programs and taught us as well how to program the machine. But he basically said that this would never work, because the products vary so much in size and type. The weeks after I took various products, set up programs in the metal detector and ran them through. Then, when I added the test pieces, the machine did not detect any of them (I had F, NF and S/S). The largest piece is a 8.0 mm S/S, and maybe it would have worked with much larger test pieces but how would you defend having anything bigger than that since it is the smaller pieces you want to find in your product    

I also tried programming the machine based on only one type of product and used marinated slices for that. That worked well and the machine detected when I added the test pieces. But....different kinds of marinade gave different results...

Based on what I have read, and the comments here as well, I guess we will have to wait with the audit until we have put up another system for this...and I'm guessing x-ray would suit our company best

[margreteva]

Hi Margreteva,

 

there is mention of knife policy & controls, and some have already commented on this point.  Is there risk of metal pieces such as lead shots, broken needles in carcass?

 

x-ray is a possible option as well.  It can have dynamic product switch (change product set up depending on which SKU is entering x-ray machine) and would be able to detect bones in boneless products as well.

 

Yes, I guess an x-ray machine would suit our company best, as it would come in handy with the boneless products as well. And as the packing room is set up, we would definitely need that dynamic product switch feature 

[Charles.C]

Last year we got a guy from a company here in Iceland which specialises in validating and verifying metal detectors. He set up a few programs and taught us as well how to program the machine. But he basically said that this would never work, because the products vary so much in size and type. The weeks after I took various products, set up programs in the metal detector and ran them through. Then, when I added the test pieces, the machine did not detect any of them (I had F, NF and S/S). The largest piece is a 8.0 mm S/S, and maybe it would have worked with much larger test pieces but how would you defend having anything bigger than that since it is the smaller pieces you want to find in your product    

I also tried programming the machine based on only one type of product and used marinated slices for that. That worked well and the machine detected when I added the test pieces. But....different kinds of marinade gave different results...

Based on what I have read, and the comments here as well, I guess we will have to wait with the audit until we have put up another system for this...and I'm guessing x-ray would suit our company best

 

This simply sounds like the MD was suffering from a  a major sensitivity failure for some reason. Time for a Service ?.

 

What was the machine specification ?

 

What was the best sensitivity achievable when you ran the test pieces on their own ?

[GMO]

Thank you for your reply!

 

The reason for this problem is that we don't really have product lines. We have a meat cutting room where the carcasses come in. There are several people working in the room cutting the meat and each one is working on a different part of the carcass. Each person then puts their product on the same conveyor belt which takes it to the packing room. There we have people putting the different products in different vacuum bags or in some cases, large laminated carton boxes. The products in the vacuum bags then go to the vacuum machine but after that they go on the same conveyor belt that takes every product past the scales, through the metal detector and then into the cold storage room. 

We have tried programming the metal detector with various products, but it doesn't work because the products vary so much.

 

 

 

Last year we got a guy from a company here in Iceland which specialises in validating and verifying metal detectors. He set up a few programs and taught us as well how to program the machine. But he basically said that this would never work, because the products vary so much in size and type. The weeks after I took various products, set up programs in the metal detector and ran them through. Then, when I added the test pieces, the machine did not detect any of them (I had F, NF and S/S). The largest piece is a 8.0 mm S/S, and maybe it would have worked with much larger test pieces but how would you defend having anything bigger than that since it is the smaller pieces you want to find in your product    

I also tried programming the machine based on only one type of product and used marinated slices for that. That worked well and the machine detected when I added the test pieces. But....different kinds of marinade gave different results...

Based on what I have read, and the comments here as well, I guess we will have to wait with the audit until we have put up another system for this...and I'm guessing x-ray would suit our company best

 

Ultimately the metal detector process isn't working.  As Charles said above there may be some issues with set up as well but as you rightly point out, different marinades can have different ingredients in them with varying molecular charges so it will give a different product signal. 

 

So for me, the problem is you're trying to run all these different products through the same metal detector without changing the settings.  It is that which is meaning you cannot get consistent metal detection.  You are trying to get a consistent result knowingly introducing a massive variable.  In my view, I'd change your process so you can metal detect all products (or at least some) on their own setting.  This may mean another metal detector or it might be something as simple as collating a quantity of finished product, e.g. 50 packs, setting up the metal detector and running them through as a batch? 

 

As an auditor I would be highly suspicious of anyone undertaking a cutting operation without metal detection, at least on a manufacturing (rather than kitchen) scale.  To try and "risk assess it away" also signals the wrong mindset to me.  It's suggesting someone who wants an easy solution rather than the right solution.  It's not the best impression to give someone coming in to look at your food safety systems.

 

Imagine if you will, a situation where you have a piece of metal that has made it to a consumer.  It causes laceration requiring significant surgery.  You are appearing in court.

Barrister "So Margerteva, you decided to stop metal detecting.  Why was that?"

Defence witness "We kept getting metal detection failures and false positives."

Barrister "Why was that?"

Defence witness "We were checking multiple products at once."

Barrister "So it would have prevented my client being injured by your products had you got a second metal detector?"
Defence witness "Well..."

Barrister "Is that possible?"

Defence witness "It's possible."

Barrister "So I put it to you that you willfully made a decision not to metal detect your product because it was easier and cheaper, is that correct?"
Defence witness "I wouldn't say that?"

Barrister "So what would you say?  I am to believe that metal detection is the industry norm?"

Defence witness "Yes that's correct."

 

Hmm.... I wouldn't like to be on the receiving end of REALLY having to defend that decision. 

[Charles.C]

Hi GMO,

 

IMO It's (initially) all about Calibration / Validation.

 

And maybe Servicing.

 

I agree that a cutting situation is a hard-to-avoid MD scenario unless this is an intermediate product where it can be documented  that the next stage guarantees detailed scanning.

 

But I anticipate that many (probably non-BRC certified) 1st-Stage meat processors don't use MDs (IIRC, I remember a (financial) caveat as to MDs not being mandatory for USDA-approved meat processors)

 

I also agree that a (fixed) MD's aperture effects may be screwing things up if the size is varying massively (as per the Service-Guy)

 

Nonetheless, a MD failing to recognize an 8mm test ball of SS on its own is surely unacceptable. And sufficient to return the MD ?

[GMO]

 

 

Nonetheless, a MD failing to recognize an 8mm test ball of SS on its own is surely unacceptable. And sufficient to return the MD ?

 

Depends on the size... a 20kg block of cheddar you will probably *just* be able to detect a 7mm SS Ballbearing.  If we're talking primals, I can imagine the situation could be similar. 

 

Do remember that the US standards on food safety are not the same as European.  We tend to be more open to discussion and risk assessment but what is acceptable in the US may not be acceptable under EU regulatory frameworks (and vice versa).

[QAGB]

I'm surprised there aren't any more folks with a meat processing background chiming in on this topic.

 

Margreteva - does your metal detector have a physical user interface so you can actually see the signal frequencies as the meat passes through - or is it one of those systems that just has a green bar for "good" and red bar for "rejected"? If you have a metal detector with a graphical interface, you can easily adjust frequencies and phase shifts to make sure you don't have false rejections, and to check your signal strength. With this type of detector, I have to think that metal detection should work. It will require several settings - maybe you cut all your carcasses as you need to, separate the parts out, and run similar pieces together under the same settings. Then switch to alternate settings as you sort through the carcasses.

[Charles.C]

Depends on the size... a 20kg block of cheddar you will probably *just* be able to detect a 7mm SS Ballbearing.  If we're talking primals, I can imagine the situation could be similar. 

 

Do remember that the US standards on food safety are not the same as European.  We tend to be more open to discussion and risk assessment but what is acceptable in the US may not be acceptable under EU regulatory frameworks (and vice versa).

 

Hi GMO,

 

Yes,  minced meat, goulash, and undefined "cuts" of meat have so far been mentioned. Sounds like some potentially interesting size-weight sensitivity variations +  MD fixed(?) aperture challenges,

 

Last year we got a guy from a company here in Iceland which specialises in validating and verifying metal detectors. He set up a few programs and taught us as well how to program the machine. But he basically said that this would never work, because the products vary so much in size and type.

Words from the Wise perhaps.

 

Unfortunately the OP has not responded over any Validatory testing results/Procedures so limited to speculations.

 

Were you suggesting that >= 8mm metal objects in their goulash might be amenable to EU consumers ?. Only in James Bond I fear -

 

 jaws.PNG   63.19KB   0 downloads