26 replies to this topic

[Mr. Incognito]

So.  I have a question.

 

The plant manager where I work has done a bunch of mock recalls and I really haven't.

 

We performed a mock recall today on an ingredient.  We found all of the product that it made and recalled 114% of the ingredient.  That part is all good.

 

But during a mock recall aren't you supposed to identify everyone you sent the affected products to?  Or just leave it how the plant manager has it and if there Is questions just pull it out during the review of the mock recall?

 

 

[Snookie]

I may not be understanding  your question, but how do you know that you identified all of the product if you don't identify where it went?  If the product was sent out to customers for example we would identify the customers in the mock recall. 

[fgjuadi]

I get copies of the BOLs of the deliveries to the customer's distribution centers and contact info for the customers if we had to contact them. Then I write fake press releases.  This is the second best part of my job, other than planning the annual fake emergency, which will be either a plane crash or some kind of time travel disaster.   Sometimes I have to wear oversized sunglasses and long white gloves and red lipstick to get into the mood, you'd be surprised at how much better you feel. 

But I was wondering how far you could take a mock recall.  Contact the distributors and see if they can track it?  I'd imgaine they have to do the same exercise. It would be a thing of beauty to see full and back forward coordinated trace and recall.  Making it a "thing to do" would just back fire into endless mock recalls due to schedules not syncing, but one would be cool. The FDA should sponsor a national Mock Recall week the GAP farmers start off.   

[cazyncymru]

Yes, you should identify all of the customers or depots that it was dispatched too and record this.

 

Caz x

[Mr. Incognito]

Thanks that's what I thought.

[Snookie]

  Sometimes I have to wear oversized sunglasses and long white gloves and red lipstick to get into the mood, you'd be surprised at how much better you feel.

  How to put the fun in recall.  

[Setanta]

I've been advised to not get much more than 100% (104-106%) because it shows that your records may not be as accurate as you think. I had a auditor really balk at our mock recall of 110%. Where did the extra come from?

[jenky]

Yes, as others have said you should be able to trace your products to the customer.  I also agree with Setanta's comment on over 100% - your tracing system should be accurate enough that you are not significantly over 100%.  In fact, we monitor all yields as products are manufactured so that if a yield is greater than 100% we can identify where it came from at the time rather than trying to figure it out weeks or months later.  Often times, the incoming bulk weight was slightly off.

 

 @ Magenta - I like the way you think.  I thought my last mock "crisis" was fun with a nefarious individual stirring up trouble, but sunglasses and other incongnito attire might have the natives paying a little more attention.  

Edited by Jennifer B., 15 May 2014 - 07:28 PM.

[AllisonB]

We use BRC for our GFSI benchmark so the actual standard may be a little different,;

 

We are required to go one step forward and one step back. I identify the suppliers where the ingredient has come from, identify all products the ingredient went into and identify the DCs where the product has shipped to  - but do not actually contact them (the contact information is noted, and I choose at random a few of the contacts to actually call to verify the information is still accurate and note as such in our closing reports.) We use an outside cold-storage facility for most of our products and I typically alert them when we will be doing our mock re-call to involve them as well, we trace to the warehouse and then they trace to the stores/DCs (then they are able to use the mock recall for their AIB audit) We are also required (during a real recall) to notify our auditing body and the FDA.

 

I also agree with the others on re-calling 114% of the product. For us- we have built into our systems allowable losses/gains for each ingredient (3-6%) Our goal is to stay within this range (because it is explainable). I had an organic auditor call us out on retracing 102% of the product. "How do you end up with more than you started with??" is what they think, which usually leads them down a rabbit trail of accuracy of records (which is a rabbit trail you want to stay far FAR away from)

[jonboy47]

So.  I have a question.

 

The plant manager where I work has done a bunch of mock recalls and I really haven't.

 

We performed a mock recall today on an ingredient.  We found all of the product that it made and recalled 114% of the ingredient.  That part is all good.

 

But during a mock recall aren't you supposed to identify everyone you sent the affected products to?  Or just leave it how the plant manager has it and if there Is questions just pull it out during the review of the mock recall?

 

Alright, I'll have a go at answering this question too.  Great topic.

 

As Allison mentioned, the general rule is "one step forward, and one step back."  If you don't trace to your immediate suppliers and your immediate customer it isn't a complete trace activity.  I would consult whatever certification you adhere to, whether a GFSI standard, or a HACCP audit of some sort, and see what it specifically requires.  Be aware that the expectation, more and more, is moving to a full-on mock recall and not just a trace exercise - meaning the activities that magenta outlined, acting as if it were a real recall (lipstick not required :-P  )

 

The question of recovery overages is a great one too.  I agree with Allison, you shoudl be able to justify your result, whatever that result is.  Anyone familiar with production operations should hopefully understand that there are variables in all processes such as waste streams, rework, etc. and that in many cases there is no such thing as precisely 100% accounting for all product. 

 

Jon

[saucy]

We also go one step forward and one step back. We require our public warehouse to furnish data on all received goods and shipment records by lot number. In addition to our trace reports, we pull all pick lists, BOLs, & invoices to the customers we ship direct, as well as ingredient and packaging specifications of the product being recalled. This is a good spot check for specs, and it would likely be required in an actual recall. We've even included specific work instructions for the product.

 

BTW, in an actual recall/withdrawal, there is often >100% recalled because many distributors or retailers won't sift through for the correct lot numbers--they simply pull all of the product.

[Tony-C]

So.  I have a question.

 

The plant manager where I work has done a bunch of mock recalls and I really haven't.

 

We performed a mock recall today on an ingredient.  We found all of the product that it made and recalled 114% of the ingredient.  That part is all good.

 

But during a mock recall aren't you supposed to identify everyone you sent the affected products to?  Or just leave it how the plant manager has it and if there Is questions just pull it out during the review of the mock recall?

 

Some sound advice already posted.

 

FSSC 22000 isn't as prescriptive as some certification schemes but here are a few relevant extracts:

 

ISO 22000 Clause 7.10.4 Withdrawals
'To enable and facilitate the complete and timely withdrawal of lots of end products which have been identified as unsafe
a) top management shall appoint personnel having the authority to initiate a withdrawal and personnel responsible for executing the withdrawal, and
b) the organization shall establish and maintain a documented procedure for
1) notification to relevant interested parties (e.g. statutory and regulatory authorities, customers and/or consumers),
2) handling of withdrawn products as well as affected lots of the products still in stock, and
3) the sequence of actions to be taken.
The organization shall verify and record the effectiveness of the withdrawal programme through the use of appropriate techniques (e.g. mock withdrawal or practice withdrawal).'

TS ISO 22002 Clause 15 Product recall procedures
'15.1 General requirements
Systems shall be in place to ensure that products failing to meet required food safety standards can be identified, located and removed from all necessary points of the supply chain.
15.2 Product recall requirements
A list of key contacts in the event of a recall shall be maintained.'

 

So to comply you should tracing to all relevant interested parties so that you can ensure that products failing to meet required food safety standards can be identified, located and removed from all necessary points of the supply chain and at the same time you should be checking your list of key contacts.

 

With regards to a guideline on an acceptable % traced 114% sounds excessive and inaccurate, an auditor may take the view that it is better to be over than under in your trace but it would concern me.

 

Regards,

 

Tony

Edited by Tony-C, 16 May 2014 - 03:06 AM.

[Sigrid B]

We did a mock recall exercise during our audit with 380 grams of product unaccounted for. The auditors 'NC-ed' it as there was no declaration in our Mock Recall Exercise Worksheet on the disposition of the missing product (whether spillage/waste from sweeping, machine hold-up, excess for tip in).  

 

Our company's Mock Recall Exercise procedure was recommended to state the percent completeness of trace that is more realistic if 100% is not feasible (due to reasons stated above) and if the result is <100%, the cause must be determined by the Mock Recall Team.  

[Tony-C]

It would be more helpful if you indicated what percentage the 380g represented.

 

Regards,

 

Tony

[cazyncymru]

http://www.bbc.co.uk...-wales-27429078

 

This might be of interest as coffee time reading!

 

Caz x

 

OOPs!

Edited by cazyncymru, 16 May 2014 - 11:19 AM.

[Tony-C]

http://www.ifsqn.com...exercise/page-2

 

This might be of interest as coffee time reading!

 

Caz x

 

Nope  

 

Do you want to try a different link?

[Oraine]

Do we needed to contact the customers or can the key contact just be distributors?

[Tony-C]

Do we needed to contact the customers or can the key contact just be distributors?

 

Hi Oraine,

 

If you deliver to distributors and not direct to customer then yes but if the product is in the market you will need to issue public notices if you have a recall

 

Regards,

 

Tony

[Jus'me]

I also must go with the majority here and state that if it is to be considered a true mock recall/trace study, you will need to determine where it went.  The one-up/one-back rule applies and of course you have to consider what is still sitting in your warehouse and what was wasted/or waiting for re-work as well as any returned product.  Sometimes that is overlooked and people are scrambling re-checking numbers on the BOL's and such and pretty well going into panic mode, until someone gets an epiphany and then all is right with the world.

I also would question such a high percentage of accountability.  114% recovery seems a bit high to me, but then it depends on the total population too, if it is a rather small amount, say 50 pieces, then 114% would be only 7 pieces extra, but if we are talking hundreds or thousands, then the recovery is a bit extreme.  Regardless, I would suggest having recovery limits to consider it a success, the closer to an actual 100% the better, and if you can maintain a fraction that is the ultimate, say 99.5-100.5% recovery in identifying the path both ways.  If you can narrow it down to that tight of a range you should never have a problem with accountability.  Again it is all statistical and as Mark Twain said there are three types of lies: Lies, damn lies and statistics. The real thing is will you be able to recover most all your product if there ever is an issue with it, if you don't know where it went to after it left your facility, then it doesn't matter how much of the raw materials you can identify, if you can't identify where it went, then it is kind of pointless.  

[Mr. Incognito]

114% really isn't high depending on the circumstances.

 

We did a mock recall on an ingredient that is metered into our final product.  We found that the ingredient probably went into the first 20 minutes of a 2 hour run therefore that entire load had to be part of the recall... and in doing so part of the next box and the rest of that product became an extra % amount of product that was recalled.

 

I think it ended around 116%.  The explanation of why we recalled 116% is that we recalled "with an abundance of caution due to the amount of material that may have been introduced into the next product run."

 

The may and could are due to the fact that we can't let the hopper run completely empty or we would have to shut down so there is always a little bit of mixing going on with the ingredients.

[Tony-C]

114% really isn't high depending on the circumstances.

 

Hi Mr. Inc..

 

You are talking about % product you would recall not % of ingredient you traced which would be from your batch records.

 

The extra % may be due to concerns of carry over (nothing wrong with being cautious) but it isn't part of a mass balance unless you know the volume/weight of carry over.

 

Normally a mass balance would consider weight/volume of product, weight/volume of batches, weight/volume of ingredient added and losses in things like overfill & fixed waste.

 

Regards,

 

Tony

[Mr. Incognito]

Sorry I don't think I was really good at conveying what I meant to say.

 

You trace the ingredient that is being recalled to the product that it made and then recall the product you made with the ingredient.  We recalled 116% of the product that had that ingredient in it because it ran into another batch of the ingredient but we aren't going to only recall the first 30 minutes of the next product run... mostly because we can't as we make bulk tankers of flour typically... so you end up with more than 100% of product recalled.

 

In fact one auditor I talked to said he looks more closely at 100% recalls because when people typically fudge numbers they try to get it to the magical 100%.

[Snookie]

Sorry I don't think I was really good at conveying what I meant to say.

 

You trace the ingredient that is being recalled to the product that it made and then recall the product you made with the ingredient.  We recalled 116% of the product that had that ingredient in it because it ran into another batch of the ingredient but we aren't going to only recall the first 30 minutes of the next product run... mostly because we can't as we make bulk tankers of flour typically... so you end up with more than 100% of product recalled.

 

In fact one auditor I talked to said he looks more closely at 100% recalls because when people typically fudge numbers they try to get it to the magical 100%.

 

 

I think you said it fairly well, but many may not understand your environment and realize that in many production facilities we do get carryover into another batch therefore you need to recall more rather than less to make sure it has all been captured.   I too would be suspicious of exactly 100% unless it was clear that was achievable accurately. 

[Tony-C]

I think you said it fairly well, but many may not understand your environment and realize that in many production facilities we do get carryover into another batch therefore you need to recall more rather than less to make sure it has all been captured.   I too would be suspicious of exactly 100% unless it was clear that was achievable accurately. 

 

That is true, you may want to recall more than the appropriate % of 100% of the ingredient as a finished product but that is not the same as a mass balance - Read my previous post again.

 

My concern here is that people are talking about different %ages:

'We performed a mock recall today on an ingredient.  We found all of the product that it made and recalled 114% of the ingredient'

 

But not 114% of the relevant ingredient batch - Get it?

 

Regards,

 

Tony

[Mr. Incognito]

No you can't recall more than 100% of the batch of ingredient but if it's an ingredient your not going to recall that ingredient you have to recall all of the product it made.  That's why when I wrote up my explanation I said we had to trace the affected ingredient to all of the product it made.

 

I think we are just talking different angles of the same thing Tony.  We are both not wrong