Jump to content

  • Quick Navigation
Photo

How to determine a recall standard?

Share this

  • You cannot start a new topic
  • Please log in to reply
10 replies to this topic

Cici2018

    Grade - Active

  • IFSQN Active
  • 15 posts
  • 0 thanks
1
Neutral

  • Canada
    Canada

Posted 25 August 2021 - 07:10 PM

How to determine a recall standard? For example, 100%, 95%, or 80% of product can be recalled means the recall is successful?



Setanta

    Grade - FIFSQN

  • IFSQN Fellow
  • 1,596 posts
  • 369 thanks
382
Excellent

  • United States
    United States
  • Gender:Female
  • Interests:Reading: historical fiction, fantasy, Sci-Fi
    Movies
    Gardening
    Birding

Posted 25 August 2021 - 07:19 PM

Generally for Mock Recalls and I would presume Real Recalls, the standard is 100%


-Setanta         

 

 

 


SQFconsultant

    SQFconsultant

  • IFSQN Fellow
  • 4,629 posts
  • 1135 thanks
1,125
Excellent

  • United States
    United States
  • Gender:Male
  • Interests:Just when I thought I was out - They pulled me back in!!!

Posted 25 August 2021 - 07:46 PM

The inability to account for anything under 100% should be viewed as a failure.


All the Best,

 

All Rights Reserved,

Without Prejudice,

Glenn Oster.

Glenn Oster Consulting, LLC -

SQF System Development | Internal Auditor Training | eConsultant

 

 

Martha's Vineyard Island, MA - Restored Republic

http://www.GCEMVI.XYZ

http://www.GlennOster.com

 

774.563.7048


TimG

    Grade - PIFSQN

  • IFSQN Principal
  • 642 posts
  • 191 thanks
320
Excellent

  • United States
    United States

Posted 25 August 2021 - 08:08 PM

Ok, I'm going to chime in on this. Your + and - is completely going to depend on your product. If you produce in bulk, there is going to have to be an allowance for a bigger swing.

In a bulk sugar facility that was run under SQF (and repeatedly audited in the 95+ range) we factored in a + or - variance of 5% from 100. We also had to factor in end of day floor sweep (discard for rework) weights so the 10%  full swing (95% to 105%) was necessary. When you're working with 10's of thousands of pounds bagged in bulk per line per day, 100% is probably not realistic.

 

I have never heard of a system allowing more than a 5% variance from 100, and I had to back up my percent error with historical data and plant procedures to the auditor before they were on board with it.



Thanked by 1 Member:

Cici2018

    Grade - Active

  • IFSQN Active
  • 15 posts
  • 0 thanks
1
Neutral

  • Canada
    Canada

Posted 25 August 2021 - 09:03 PM

Thank you for your reply. Do yo have some reference for making the standard? I want to add some references in our recall SOP.

 

And for determining the recall percentage, TimG is correct. For example, we are using some seasoning ingredient to produce seasoned meat, is there is a recall for this seasoning ingredient. We need to recall all product we produced using that seasoning ingredient and also the seasoning ingredient we have in storage and we shipped out. We contact with customer, we get some of the products back, the rest of them have been consumed. For the seasoning ingredient, we know how many kg we received, how many kg are used, and how many kg are left in storage. But the total recalled  weight is less than the total weight we received, because during production, some of the ingredients were wasted. If the variance is >5%, is that a failure?



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 26 August 2021 - 01:17 AM

How to determine a recall standard?  For example, 100%, 95%, or 80% of product can be recalled means the recall is successful?

 

Hi Cici,

 

^^^^ (red) - Cannot see any reference to this (curious) terminology in BRC Food Standard ?

Presumably Clause 3.9.3

Not sure if referring to forward traceability or backward (ca. mass balance).  Strictly BRC seems to include both  but Posts 1/ 5 seem to relate to the former/latter respectively ?

 

One numerical example of former here -

 

https://www.ifsqn.co...ort/#entry47832


Edited by Charles.C, 26 August 2021 - 02:29 PM.
edited

Kind Regards,

 

Charles.C


Venkateshkini

    Grade - AIFSQN

  • IFSQN Associate
  • 37 posts
  • 3 thanks
6
Neutral

  • India
    India

Posted 26 August 2021 - 06:03 AM

Generally recovered product shall be 100%. But if product is already consumed then it shall be 100% of left over product in retailers and wholesalers. 



TimG

    Grade - PIFSQN

  • IFSQN Principal
  • 642 posts
  • 191 thanks
320
Excellent

  • United States
    United States

Posted 26 August 2021 - 01:36 PM

Thank you for your reply. Do yo have some reference for making the standard? I want to add some references in our recall SOP.

 

If the variance is >5%, is that a failure?

If I had gone outside my set limits, then yes the trace exercise would have been a failure. It's been 8 years since I've been in that facility, but I still have access to the old files so I just had to go look at them to refresh. Prior to streamlining we had target goal of 99.5-100.5%, minor NC for anything in the 95-99.4% 100.5-105.0 (+or-5%), major NC for any 5.1-10% swing, and failure for anything over 10% from 100 (so <89.9 or >110.1). The problem with that was 70% of our mock recalls left us in a minor NC (we would do them every month if we had a minor). We only had the accuracy to get it reliably within 95-105. It was re-written so that anything within the 95-105 was a pass and anything out was a fail; we better aligned the procedure while also tightening it up. To me the minor/major/fail was too loose and only a safety net to make things look better. I had the historical info to back up what our accuracy was.

 

My suggestions:

  • make sure BRC and CFIA don't have any hard and fast accountability requirements for recall (as I said, I was under SQF/FDA)
  • determine what your acceptable percent error is going to be, and then show WHY on paper, perhaps even explaining in your trace exercise PnP
    • you might even find that you can tighten up your procedures to lower this value
  • Run them every month for a few months, gathering more data (perhaps tightening it up even more)


Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 26 August 2021 - 02:33 PM

 

If I had gone outside my set limits, then yes the trace exercise would have been a failure. It's been 8 years since I've been in that facility, but I still have access to the old files so I just had to go look at them to refresh. Prior to streamlining we had target goal of 99.5-100.5%, minor NC for anything in the 95-99.4% 100.5-105.0 (+or-5%), major NC for any 5.1-10% swing, and failure for anything over 10% from 100 (so <89.9 or >110.1). The problem with that was 70% of our mock recalls left us in a minor NC (we would do them every month if we had a minor). We only had the accuracy to get it reliably within 95-105. It was re-written so that anything within the 95-105 was a pass and anything out was a fail; we better aligned the procedure while also tightening it up. To me the minor/major/fail was too loose and only a safety net to make things look better. I had the historical info to back up what our accuracy was.

 

My suggestions:

  • make sure BRC and CFIA don't have any hard and fast accountability requirements for recall (as I said, I was under SQF/FDA)
  • determine what your acceptable percent error is going to be, and then show WHY on paper, perhaps even explaining in your trace exercise PnP
    • you might even find that you can tighten up your procedures to lower this value
  • Run them every month for a few months, gathering more data (perhaps tightening it up even more)

 

 

Hi Tim,

 

Apologies if I have misunderstood yr Posts 4 / 8 but content looks like a (raw material) backwards traceability/mass balance exercise ?.

 

Can compare to the (basic) accounting exercise described in the  "Product Recall Report" in  my link of Post 6 (and surrounding posts in same thread).


Kind Regards,

 

Charles.C


TimG

    Grade - PIFSQN

  • IFSQN Principal
  • 642 posts
  • 191 thanks
320
Excellent

  • United States
    United States

Posted 26 August 2021 - 03:40 PM

Sorry Charles, I guess I didn't clarify.

My posts are entirely dealing with final product produced by one facility and bagged by the other.

Incoming unrefined sugar is a commodity and tracked by the vessel, it was also not considered a food product until we turned it into one. Our facility was the beginning of the FDA regulated food chain.



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 26 August 2021 - 10:58 PM

Sorry Charles, I guess I didn't clarify.

My posts are entirely dealing with final product produced by one facility and bagged by the other.

Incoming unrefined sugar is a commodity and tracked by the vessel, it was also not considered a food product until we turned it into one. Our facility was the beginning of the FDA regulated food chain.

 

Hi Tim,

 

Thks for clarification.

 

There is often confusion in the use of terminologies traceability/mock recall/product recall/mass balance.

 

As I understand, your percentages in Post 8 were referring to a mass balance for the bagging operation. This seems to be a "special" interpretation of (FS) Mock Recall.

 

Here is BRC's overlaying Guideline -

 

In keeping with the traceability requirements for manufactured and processed products, sites that trade food products are required
to maintain traceability [backward/forward], allowing food businesses and authorities to withdraw or recall products that have been identified as unsafe.

 

BRC Standard makes no specific mention of "mock recall" but it represents a Product Recall "methodology" involving, inter alia, quantitative  forward traceability (eg see previous link). Demonstration of Backward traceability can be quantitated via a mass balance.

 

@ Cici - IMO the answer to yr Posts 1, 5 is Probably, mostly, NO (assuming you can validate the result for post 5). Full traceability is stated as typically expected within 4 hours (with caveats). (the BRC,  I.G. is a "Guideline").

 

I suggest you obtain the BRC Interpretation Guide . This has a substantial discussion of the accuracy expectations concerning product recall/forward/backward traceability/ mass balance (eg sections 3.9, 9.5).

 

.


Kind Regards,

 

Charles.C




Share this


0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users