Jump to content

  • Quick Navigation
Photo

Metal Detector Not Verified During Operation

Share this

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

jenky

    Grade - MIFSQN

  • IFSQN Member
  • 77 posts
  • 37 thanks
13
Good

  • United States
    United States
  • Gender:Female

Posted 15 February 2018 - 02:42 PM

I am hoping to get some feedback on a recent incident with our metal detector verification check.  We have an in-line metal detector installed shortly before the filler.  Metal detection is a CCP - one verification activity is that the detector is challenged with test pieces every two hours. 

 

We had a incident where the metal detector was not challenged for 8 hours.  We know why and are addressing that issue.  The problem is how to disposition the 8 hours of product.  Our QA team feels all of the product between the last good check and the check that was done immediately when the problem was found should be reworked.  Our production team is challenging this since the metal detector was operating, even though we did not check it.  The last good check and the check at 8 hours when the problem was discovered both show that the detector was operating properly.  Thoughts? 



FSQA

    Grade - SIFSQN

  • IFSQN Senior
  • 315 posts
  • 126 thanks
55
Excellent

  • United States
    United States

Posted 15 February 2018 - 02:57 PM

Hi Jenky,

 

Just my 2 cents,

 

I would recommend to stick with your QA's decision to rework all the product that was processed from the last check till the error was detected.

As you have Metal detection as a CCP and its set for a 2 hour verification time frame, I am assuming the corrective actions (in your HACCP) for a CCP verification failure would also be dictating to rework the product.

 

Practically, it seems like the Metal Detector was working between the 8 hour window, however, the 2 hour window that your HACCP team has selected for verification, is based on the Hazard Analysis.

 

What if the metal detector failed for sometime during the 8 hour window? Is someone from production taking the responsibility that if something goes wrong they will be held liable?

 

Again, these are just my 2 cents, i am pretty sure we have much more experienced individuals than me over here and can share their experienced points on this issue.



Thanked by 1 Member:

BrummyJim

    Grade - SIFSQN

  • IFSQN Senior
  • 329 posts
  • 117 thanks
26
Excellent

  • United Kingdom
    United Kingdom
  • Gender:Male
  • Location:South West
  • Interests:Motorbike gone now. Only the dog to walk!

Posted 15 February 2018 - 03:16 PM

I'm with FSQA here. Rework it all from when the check was due. Your HACCP plan should tell you this anyway. Would you feel the same way if it were the pasteuriser, and what would you say to your auditor?



Thanked by 1 Member:

itreatpets

    Grade - MIFSQN

  • IFSQN Member
  • 73 posts
  • 13 thanks
4
Neutral

  • Canada
    Canada

Posted 15 February 2018 - 03:28 PM

If your packaging does not interfere with your metal detector functioning properly, could you pass the product through the metal detector still in the package?

 

This may be an option where you address this issue but also still working with the production team. If not, I recommend the rework. Better safe than sorry.



Thanked by 1 Member:

jenky

    Grade - MIFSQN

  • IFSQN Member
  • 77 posts
  • 37 thanks
13
Good

  • United States
    United States
  • Gender:Female

Posted 15 February 2018 - 03:35 PM

Unfortunately, our product is packed in bulk drums so it can be passed through the metal detector completely reworking it. 



FurFarmandFork

    Food Safety Consultant, Production Supervisor

  • IFSQN Fellow
  • 1,264 posts
  • 590 thanks
206
Excellent

  • United States
    United States
  • Gender:Male
  • Location:Oregon, USA

Posted 15 February 2018 - 03:45 PM

the 2 hour window that your HACCP team has selected for verification, is based on the Hazard Analysis.

 

 

If this is the case, I agree with the above, rework according to your HACCP plan, if you plan says it, you have to do it. That's why everyone signs off on it. :) If they want to change it for the future do it as a preventive action.

 

HOWEVER, monitoring intervals aren't always based on the hazard analysis, they can also be based on scope of rework. For example, I have a check on our production line that relies on chemical dosing, and I have high confidence (based on years of data) that it rarely goes out of spec and needs little adjustment with an experienced operator.

 

Based on that hazard analysis, I potentially only need to monitor the status at startup, after breaks, and shutdown. However my only corrective action available if that check fails is to rework all product since last good check. So while I have confidence in the equipment, we instead check it every hour so that in the event of a failure we only need to rework one hour of product instead of 4-6 hours.

 

My corrective actions in my HACCP plan reflect this, so while the actual monitoring in the plan is stated for startup and shutdown, the daily production paperwork has my guys checking it every hour to minimize the scope of rework in the event of a failure.

 

If you feel good about your metal detector, you can revisit your HACCP plan and do the same thing without changing your practices on the line. However if your plan doesn't say it now, you're stuck with the plan as written, otherwise what's the point of the plan? Say what you do, do what you say.

 

Metal detection checks are often driven by customer requirements however, rather than your particular hazard analysis, so you need to make sure you include those in your assessment.


Austin Bouck
Owner/Consultant at Fur, Farm, and Fork.
Consulting for companies needing effective, lean food safety systems and solutions.

Subscribe to the blog at furfarmandfork.com for food safety research, insights, and analysis.

Thanked by 4 Members:

Angela5555

    Grade - Active

  • IFSQN Active
  • 9 posts
  • 2 thanks
0
Neutral

  • Earth
    Earth

Posted 16 February 2018 - 12:02 PM

All product from the last successful check should be reworked and recorded! I can only assume that your records have a 8 hrs gap hence incomplete.  



vijesh2002

    Grade - Active

  • IFSQN Active
  • 5 posts
  • 23 thanks
2
Neutral

  • India
    India

Posted 16 February 2018 - 12:32 PM

I will stick with your QA decision since this is based as per your HACCP plan.

Bit deviating from topic, is it right to keep metal detection as CCP since it is not eliminating 100% metals. It can be made OPRP. Any comments pls



FurFarmandFork

    Food Safety Consultant, Production Supervisor

  • IFSQN Fellow
  • 1,264 posts
  • 590 thanks
206
Excellent

  • United States
    United States
  • Gender:Male
  • Location:Oregon, USA

Posted 16 February 2018 - 03:42 PM

 

Bit deviating from topic, is it right to keep metal detection as CCP since it is not eliminating 100% metals. It can be made OPRP. Any comments pls

 

That's a whole different can of worms vijesh2002, answer is potentially depending on your country, customer, and role of metal detection in the safety of your product.

 

Discussions can be found here:

http://www.ifsqn.com...ivity-or-a-ccp/

http://www.ifsqn.com...-and-not-a-ccp/

http://www.ifsqn.com...metal-detector/


Austin Bouck
Owner/Consultant at Fur, Farm, and Fork.
Consulting for companies needing effective, lean food safety systems and solutions.

Subscribe to the blog at furfarmandfork.com for food safety research, insights, and analysis.

Thanked by 1 Member:

Ryan M.

    Grade - FIFSQN

  • IFSQN Fellow
  • 1,329 posts
  • 479 thanks
290
Excellent

  • United States
    United States
  • Gender:Male
  • Location:Birmingham, AL
  • Interests:Reading, crosswords, passionate discussions, laughing at US politics.

Posted 16 February 2018 - 04:45 PM

Look at your history of verification testing.  If your history is showing checks are consistently good in an 8 hour period you have a definite case to keep the product as is given the history and the testing verified for that production day in question.

 

If your history is suspect, then rework the product.



Thanked by 1 Member:

FSQA

    Grade - SIFSQN

  • IFSQN Senior
  • 315 posts
  • 126 thanks
55
Excellent

  • United States
    United States

Posted 16 February 2018 - 05:05 PM

Look at your history of verification testing.  If your history is showing checks are consistently good in an 8 hour period you have a definite case to keep the product as is given the history and the testing verified for that production day in question.

 

If your history is suspect, then rework the product.

 

If results are consistent shouldn't they be revisiting their HACCP plan, to change the verification checks to 8 hours instead of 2 hours?



Ryan M.

    Grade - FIFSQN

  • IFSQN Fellow
  • 1,329 posts
  • 479 thanks
290
Excellent

  • United States
    United States
  • Gender:Male
  • Location:Birmingham, AL
  • Interests:Reading, crosswords, passionate discussions, laughing at US politics.

Posted 16 February 2018 - 05:15 PM

If results are consistent shouldn't they be revisiting their HACCP plan, to change the verification checks to 8 hours instead of 2 hours?

 

Ah!  The answer to that is..."It Depends".  Why?  What level of risk they are willing to have with the process.  If they did change it to 8 hours it may not be worth the risk to monitor only every 8 hours versus the time it takes to conduct the verification and the product output during that time.



amorable

    Grade - Active

  • IFSQN Active
  • 7 posts
  • 0 thanks
0
Neutral

  • Earth
    Earth

Posted 12 April 2018 - 05:57 PM

FurFarmandFork, why do you have it written as being at start-up and shutdown if you're checking it every hour anyway? It this to prevent the need for re-work if an hourly check is missed?



Scampi

    Fellow

  • IFSQN Fellow
  • 5,492 posts
  • 1512 thanks
1,553
Excellent

  • Canada
    Canada
  • Gender:Not Telling

Posted 12 April 2018 - 06:33 PM

I'm with FFF.....CCP's must be checked at start up and end of day....otherwise how do you know for sure it was functioning properly ALL day. Say you don't check first thing prior to running product and your checks are done every 2 hours, then you realize at that check it isn't working.....you could have caught the problem prior to running anything...rework is very expensive. The HACCP plan should also save the company money in the long run

 

If your risk says you only need to check you detector every 8 hours, why use one at all? Then it begs the question, how valid is the risk assessment?


Please stop referring to me as Sir/sirs


FurFarmandFork

    Food Safety Consultant, Production Supervisor

  • IFSQN Fellow
  • 1,264 posts
  • 590 thanks
206
Excellent

  • United States
    United States
  • Gender:Male
  • Location:Oregon, USA

Posted 12 April 2018 - 09:32 PM

FurFarmandFork, why do you have it written as being at start-up and shutdown if you're checking it every hour anyway? It this to prevent the need for re-work if an hourly check is missed?

 

So the check is for detergent concentration on a washing line. It's from an automated doser (no feedback loop, timer only) that slowly feeds soap and fresh water in as materials are lost or screened out.

This means that without adjustment, there's either extra soap (doesn't violate the CCP, crazy high amounts would break the machine), or it slowly would lose concentration throughout the day by not dosing enough for the soil load it was handling (verified via historical data).

 

A check at the beginning of the day ensures concentration starts high, a check at the end of the day would verify that it stayed high. To ensure that the day's production was safe to sell, I would only need to check beginning and end of shift, hence the CCP.

 

However, if I have a failure at the end of the day, then I have to dump and rework all products produced throughout the entire day since I have no way to limit scope. So the monitoring interval in the HACCP plan which exists to make sure the food is safe and release the product requires only those shift checks. However, we check every hour so that in the event the concentration falls below the critical limit, we can limit the scope of rework back to the last good production check, rather than the entire day's labor. If a monitoring event was missed, that isn't a CCP failure, we would just have to deal with the consequences of a failure.

 

A similar example is batching supplier raw materials. You can maintain tracebility but your scope of recall would be larger. FDA uses the language "operational limits" for this type of thing. If you had a continuous oven baking cookies and your CCP was that the oven is maintained at 350ºf, that's your CCP. But you may tell your operators to monitor temperature and not to run if it falls under 352ºf. That way if they have a check where it was 351º, they would stop and fix the problem, but you wouldn't have to throw away all the product for a CCP failure.


Austin Bouck
Owner/Consultant at Fur, Farm, and Fork.
Consulting for companies needing effective, lean food safety systems and solutions.

Subscribe to the blog at furfarmandfork.com for food safety research, insights, and analysis.

amorable

    Grade - Active

  • IFSQN Active
  • 7 posts
  • 0 thanks
0
Neutral

  • Earth
    Earth

Posted 12 April 2018 - 10:43 PM

I see. I guess I was wondering more so why you wouldn't write it as a check every hour since that is what you do already but as you said if a check is missed you wouldn't necessarily fail, you just deal with the 2 hours worth of re-work if the next check fails. I'm new to food and wanted to make sure I understood the reasoning.

Thanks for the clarity!



redfox

    Grade - SIFSQN

  • IFSQN Senior
  • 481 posts
  • 163 thanks
24
Excellent

  • Philippines
    Philippines

Posted 13 April 2018 - 01:55 AM

Hello,

 

IMEX, metal detector will not malfunction and will back to normal without any adjustment ex. phasing and sensitvity. IMO, if on eighth hour that it is still functioning well, it means, on that 8-hour span the MD did its function. We in our plant, are doing every 30 mins, so that if there is MD malfunction we only have few lbs of product in question.

 

I also point out jenky, that you lack or to loose in supervision to check if your operator has done his job in an 2-hour MD verification. It wouldn't go that far if somebody diligently check the work of your operator if your operator misses his scheduled MD verification. There is should be a "guard to guard the guard".

 

regards,

redfox



FSQA

    Grade - SIFSQN

  • IFSQN Senior
  • 315 posts
  • 126 thanks
55
Excellent

  • United States
    United States

Posted 13 April 2018 - 12:51 PM

IMO, it all depends on the risk of product re-working that one is to avail. Few Bigger companies might have the liberty to assign people just for this verification, while in few small companies the same person might be assigned to various job functions making it difficult to check at very brief intervals.

 

It basically depends on the Management of a company/operating site to determine how much of a risk are they willing to take in case of a Metal detector test failure/ 30mins/2hrs/4hrs checks (with keeping in mind that product re-work can be costly), which can be justified by a hazard/risk assessment and a metal detector history.

 

However, I m with FFF on this...CCP's must be checked at start up and end of day to verify the detector was functional at the beginning of the run and it was functional at the end of the shift to verify the processed product is not non-compliant.



FurFarmandFork

    Food Safety Consultant, Production Supervisor

  • IFSQN Fellow
  • 1,264 posts
  • 590 thanks
206
Excellent

  • United States
    United States
  • Gender:Male
  • Location:Oregon, USA

Posted 13 April 2018 - 02:30 PM

I see. I guess I was wondering more so why you wouldn't write it as a check every hour since that is what you do already but as you said if a check is missed you wouldn't necessarily fail, you just deal with the 2 hours worth of re-work if the next check fails. I'm new to food and wanted to make sure I understood the reasoning.

Thanks for the clarity!

Yep. Remember that if you fail to follow your monitoring interval that can be a CCP failure also (e.g. if you know that your temperature can go out of specification for a period of high load, but then come back in). So don't establish monitoring intervals in your HACCP plan that you don't have corrective actions for. Hence the OP.

 

@FSQA, thanks! Though it's worth noting that I disagree that the detector can't go out then back in specification, it's certainly possible, it depends on whether the historical data for your MD substantiates the monitoring interval or not.


Austin Bouck
Owner/Consultant at Fur, Farm, and Fork.
Consulting for companies needing effective, lean food safety systems and solutions.

Subscribe to the blog at furfarmandfork.com for food safety research, insights, and analysis.

Thanked by 1 Member:

fstRicky

    Grade - AIFSQN

  • IFSQN Associate
  • 37 posts
  • 10 thanks
1
Neutral
  • Gender:Male
  • Location:Down under

Posted 20 June 2018 - 07:11 AM

I too agree with FFF, the issue of time interval is mainly to determine how much product will go on HOLD (rework) in case of a failed test. A shorter time means less product a vice versa.

But you will need a good explanation in an Audit to back your course of action - i.e. if you don't follow what the plan stipulates.



012117

    Grade - MIFSQN

  • IFSQN Member
  • 227 posts
  • 69 thanks
36
Excellent

  • Earth
    Earth
  • Gender:Male
  • Location:Philippines
  • Interests:Validation, basketball, chocolatier

Posted 24 July 2018 - 09:10 AM

Follow on what is in your HACCP plan. If the process is assessed properly and there is a robust validation of the HACCP plan, then without exception, the plan must be followed. If this plan is agreed by the affected heads, the action in case of deviation should not be in question in that moment. What can be done is that, review the study and check if there are factors being missed in the validation that may require you to change the plan.

 

Hello,

 

IMEX, metal detector will not malfunction and will back to normal without any adjustment ex. phasing and sensitvity. IMO, if on eighth hour that it is still functioning well, it means, on that 8-hour span the MD did its function. We in our plant, are doing every 30 mins, so that if there is MD malfunction we only have few lbs of product in question.

 

I also point out jenky, that you lack or to loose in supervision to check if your operator has done his job in an 2-hour MD verification. It wouldn't go that far if somebody diligently check the work of your operator if your operator misses his scheduled MD verification. There is should be a "guard to guard the guard".

 

regards,

redfox

While in theory, we should expect the metal detector to work as planned if properly validated, there are instances where it may fail the work due to some factors in environment. If that is the case that you see your metal detector  have a low risk of not working? Why not reduce the testing to every 4 hours or 8 hours? While the purpose of HACCP is to provide you "confidence" that some concept of your food safety management system is working, indirectly, if with proper validation, it must also help you to reduce the unnecessary test provided you properly assess the risk.





Share this


Also tagged with one or more of these keywords: metal detector verification, CCP verification, corrective action

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users