30 replies to this topic

[Chac]

Hello everyone!

In our production line we have three filtersystems connected afterwards. In our opinion this is not a CCP because the chance that all three filter fell out is quite low.
Now we have an auditor here who of course has a different opinion about this.

What do you think?

[Charles.C]

Dear Chac,

Why is this related to product safety, ie what is the specific health hazard in the (unknown) process ?

Rgds / Charles.C

[Chac]

We're producing liquid sugar and the food safety aspect is the foreign body management. After finishing producing we are checking the filter no1. If the filter is ok we are just going on because foreing body are hold back within this step. If the filter is damaged we are checking filter no2 and go on in the same way like we did with the first filter. Only in case the third filter is damaged too we are stopping the charge.
But how likely is that??

[Sanjay INDANI]

Dear Chac,

First of all you must know that whether this hazard is SIGNIFICANT or not. If it is, then you apply the Codex decision tree for this significant hazard. By applying this you will get your third sieve as a CCP. First one can be oPRP.

Regards,

Sanjay.

[Tony-C]

Hello everyone!

In our production line we have three filtersystems connected afterwards. In our opinion this is not a CCP because the chance that all three filter fell out is quite low.
Now we have an auditor here who of course has a different opinion about this.

What do you think?

Hi Chac

Why do you have 3 filters? and are they in series?

Assuming they are in series then normally the last one would be a CCP and the ones before may be OPRP's and so all should be checked but perhaps at different intervals. If in your system you are only checking the first one and the last one is the CCP then your auditor may raise this as a non conformance.

Regards,

Tony

[Chac]

The three filters are in series but the last one is only used for clarification ( so really small particles) but it also can held back bigger particles. I've thougt that the last filter would be the CCP but it is only checked in the case that the other two filter fell out, which of course will not happen that much.

So can this be a CCP? Are all three filters a CCP?

I'm totally confused

[Tony-C]

The three filters are in series but the last one is only used for clarification ( so really small particles) but it also can held back bigger particles. I've thougt that the last filter would be the CCP but it is only checked in the case that the other two filter fell out, which of course will not happen that much.

So can this be a CCP? Are all three filters a CCP?

I'm totally confused

The last filter is your CCP and should be checked regularly. You should always be carrying out checks on your CCP's.

By your e-mail I guess they are different sizes and reduce in size. In this case they should all be checked but not necessarily at the same frequency. The first 2 filters could be regarded as Good Manufacturing Practice or Prerequisite Programmes and checked based on risk assessment.

Regards,

Tony

[Charles Chew]

The first 2 filters could be regarded as Good Manufacturing Practice or Prerequisite Programmes and checked based on risk assessment.

Agree assuming that the porosity size of all the filters are similar or reducing. While the last filter is monitored under the HACCP Plan as a CCP, the first two filters are monitored as OPRP while all filters MUST be risk assessed and monitored at the same frequency to ensure the last filter is not subject to unnecessary process stress leading it to unexpected reduction in operating shelf life. It would be grave error to suggest that the status of the control may relieve the degree of frequency in monitoring.

Regards
Charles Chew

[Tony-C]

Agree assuming that the porosity size of all the filters are similar or reducing. While the last filter is monitored under the HACCP Plan as a CCP, the first two filters are monitored as OPRP while all filters MUST be risk assessed and monitored at the same frequency to ensure the last filter is not subject to unnecessary process stress leading it to unexpected reduction in operating shelf life. It would be grave error to suggest that the status of the control may relieve the degree of frequency in monitoring.

Regards
Charles Chew

Hi Charles

Thanks for your post.

I think you are agreeing with me but I am not clear on some parts.

What has operating shelf life got to do with foreign body hazards?

What exactly do you mean by "It would be a grave error to suggest that the status of control may relieve the degree of frequency of monitoring"

Regards,

Tony

[Charles.C]

Dear All,

Perhaps a minor point. This is not the ISO 22000 forum. Perhaps Chac has no idea as to meaning of "OPRP"?

(BTW, if we were talking about ISO 22000, my offhand interpretation wud be that, if monitoring not possible to be continuous then OPRP, or if possible (?) this being a possibly synergistic situation (eg see my ccp tree - http://www.ifsqn.com...showt...&st=0 , post16, last filter might not be a CCP (if first 2 failed, cud the last one still completely take over ??). And if the health hazard is anyway minimal [is it ??] might not even be an OPRP ). Of course if it were a metal detector, > different scenario, I presume yr hazard is not metallic (foreign body = ???). To put it another way, not quite sufficient data. )

If ISO 22000 not involved, the standard Codex tree suggests that if you believe that there is a genuine significant hazard and that the third filter is required, intended and able to achieve the control of the hazard a CCP exists at the last point. This appears to require that the first 2 filters are not considered sufficient to achieve the necessary result, eg for sizing reasons. Yr first post suggests last filter really only there for non-safety purposes (clarification - unless a health regulatory std involved ??) in which case it would be irrelevant to the hazard analysis. So what is the actual situation ?? Presumably a simple risk assessment (Likelihood x Severity) wud give a similar result.

Chac, sorry if you are still confused but, as someone pointed out already, it comes down to yr (validatable) risk analysis which you hv not yet suggested. Do you hv one ? Or was that the real intention of yr original query ?

Rgds / Charles.C

[Tony-C]

Perhaps a minor point. This is not the ISO 22000 forum. Perhaps Chac has no idea as to meaning of "OPRP"?

Yes and we can get a bit hung up on 22000.

Yr first post suggests last filter really only there for non-safety purposes (clarification - unless a health regulatory std involved ??)

Possibly but would it still be a CCP if you use the standard decision tree?

Regards,

Tony

[Charles.C]

Dear Chac,

I assume for the below that you are not interested in ISO 22000 aspects (???). If otherwise, things may (probably will) change. I am still unsure as to yr precise hazard and its risk significance .

As you can see you hv presented a question with various responses. Congratulations ! There is some similarity in yr query to earlier threads here, particularly relating to screening in grain systems. For a few examples (including combining screens, filters / audit acceptability) can see these threads (some occasional overlaps to ISO 22000 ) –

http://www.ifsqn.com...showtopic=12308

http://www.ifsqn.com...showtopic=10486

http://www.ifsqn.com...showtopic=13415

http://www.ifsqn.com...?showtopic=9986

Clearly there are a (wide) range of opinions and probably IMO not just one definite, presentable, answer for the general auditor’s view. I believe you can also find different opinions in published documents (and auditors!). The surrounding situation (eg hazard analysis, monitoring) in yr specific case will be very important.

I extracted some text from IT to the attachment below which may help you to clear any conceptual confusion (some of the statements are not automatically evident from a codex tree IMO). If still confused, some more feedback is probably required to progress much further ??

 CCP_vs_CP.png   198.49KB   16 downloads

I think one important question for you is whether each filter can work independently to achieve the desired result. If not then some synergy is surely involved (and as Tony stated, some monitoring!). In some cases I expect this synergy is intentional, eg use of reducing screen mesh sizes ( IMEX smaller mesh screens are relatively easily broken by, for example, large stones hence the progression of mesh size.) However I wud not imagine that any sizeable objects are existing in yr system (?)

Rgds / Charles.C

[Biss]

Hi,

How a filter can be treated as a CCP ? how can you fix a critical limit for filter. i think it should be a OPRP.

[Tony-C]

Hi, How a filter can be treated as a CCP ? how can you fix a critical limit for filter. i think it should be a OPRP.

Hi Biss

It can and it is in many industries

Critical limit - is it intact and is it in place
Some people also have a fixed pressure through the filter

Monitoring - Check at start and end of production

Corrective Action - Hold production and investigate

Is it the last stage in a process where a hazard is eliminated - Yes it removes foreign bodies

I would consider a pre-filter a OPRP

It seems to me too many people on the forums are getting hung up on what are CCP's and OPRP's

If you are not really sure then my advice is forget about OPRP's and PRP's and consider your HACCP system in terms of Critical Control Points and Good Manufacturing practices.

Kind regards,

Tony

Edited by Tony-C, 17 February 2010 - 02:36 PM.

[Chac]

If you are not really sure then my advice is forget about OPRP's and PRP's and consider your HACCP system in terms of Critical Control Points and Good Manufacturing practices.

That's easy. As Charles already figured out this isn't about ISO 22000, in fact I don't know anything about this standard (yet).

@ Charles: Thanks for the links. That was a lot to read but it was helpful.

I consider the second filter as a CCP. It's the last step in the processline to rduce foreign bodys (which includes also metall, though we don't have a metall detector) as the third filter is only used for clarification.

BTW: Maybe it's a stupi question an I proof myself as a totally beginner but: What is IMEX??

[Charles.C]

Dear Chac,

I'm happy you were able to get to a conclusion.

IMEX = In MY Experience, you are right, it's not exactly global-friendly

Out of curiosity, are the first 2 filters identical mesh size and does the first one normally catch everything or some go to the second one also ??

I am also curious as to how you will (size) justify the "foreign bodies" on first 2? filters are a significant health hazard but not the third one.

Rgds / Charles.C

[Biss]

Hi tony,

as per the ISO 22000 standard 7.6.3 'Critical limits shall be measurable'.

Also as per the ASTA guidelines about critical limit says

'Critical limits are the boundaries of safety for preventive measures put in place at CCPs.
A critical limit will usually be a reading or observation such as temperature, time, pH,
etc. A critical limit can be an upper limit where a set amount or level cannot be exceeded.
A critical limit can also be a lower limit where a minimum amount is required to produce
the safe effect'

I think it is difficult to consider intactness as a critical limit

how can we justify

what others think !

Biss

[Tony-C]

Hi tony,

as per the ISO 22000 standard 7.6.3 'Critical limits shall be measurable'.

Also as per the ASTA guidelines about critical limit says

'Critical limits are the boundaries of safety for preventive measures put in place at CCPs.
A critical limit will usually be a reading or observation such as temperature, time, pH,
etc. A critical limit can be an upper limit where a set amount or level cannot be exceeded.
A critical limit can also be a lower limit where a minimum amount is required to produce
the safe effect'

I think it is difficult to consider intactness as a critical limit

how can we justify

what others think !

Biss

Dear Biss,

Thanks for your response.

Firstly this is not a 22000 forum

Secondly I fail to see the relevance of the ASTA Guidelines unless it covers "Sugar and Spice and all things Nice"

CODEX is regarded as the bible for HACCP Principles and so I'll quote from it:

Critical limits must be specified and validated for each Critical Control Point. In some cases more than one critical limit will be elaborated at a particular step. Criteria often used include measurements of temperature, time, moisture level, pH, Aw, available chlorine, and sensory parameters such as visual appearance and texture.

Have a think about the process in question and let me know if you still think that filtration isn't a CCP at some stage.

Kind regards,

Tony

[Biss]

thanks tony

[Charles.C]

At the risk of having to avoid a missile, I did notice that the English version of ISO 22000, 7.6.3 also contains the sentence - "Critical limits based on subjective data (such as visual inspection of product, process, handling etc) shall be supported by instructions or specifications and/or education and training."

I suppose it depends on one's definition of "measurable"

Nonetheless it does seem likely to me that such sensory monitoring would typically be undeniably discontinuous, some reasonably authoritative interpretations of the ISO 22000 standard such as Procert would then classify this as an OPRP.

This opportunity for distinction is perhaps one of the (few) clevernesses of 22000

For traditional HACCP, it is also possible to find negative opinions on non-continuous monitoring. Codex likes to try and accommodate all sides where possible. I think the usual escape clause is that the monitoring frequency should be able to be validated as not jeopardising the ability for appropriate corrective action. Whatever that means.

What a subjective world we live in.

Rgds / Charles.C

[Biss]

thanks charles for your valuable feedback.

[Tony-C]

At the risk of having to avoid a missile, I did notice that the English version of ISO 22000, 7.6.3 also contains the sentence - "Critical limits based on subjective data (such as visual inspection of product, process, handling etc) shall be supported by instructions or specifications and/or education and training."

I suppose it depends on one's definition of "measurable"

Nonetheless it does seem likely to me that such sensory monitoring would typically be undeniably discontinuous, some reasonably authoritative interpretations of the ISO 22000 standard such as Procert would then classify this as an OPRP.
This opportunity for distinction is perhaps one of the (few) clevernesses of 22000
For traditional HACCP, it is also possible to find negative opinions on non-continuous monitoring. Codex likes to try and accommodate all sides where possible. I think the usual escape clause is that the monitoring frequency should be able to be validated as not jeopardising the ability for appropriate corrective action. Whatever that means.
What a subjective world we live in.
Rgds / Charles.C

Dear Charles

Tut tut tut - A missile no

A cheque from ISO for continuous promotion of ISO 22000 maybe

All I am asking people is forget about the terminology, forget about 22000 ask yourself if using the decision tree (Codex) filtration would be a CCP or not.

Kind regards,

Tony

Edited by Tony-C, 18 February 2010 - 06:00 PM.

[Jon5]

Chac-
For my own information, are you actually filtering concentrate? I assume you're filtering juice prior to the concentration step? I can't imagine filtering such a viscous product.

Tony -
Precisely. Don't over-complicate things. Just use the tree. In this case, the step IS designed to remove or reduce a hazard. The hazard COULD occur in excess of acceptable levels. And from what I understand, no subsequent step is in place after these filters to reduce the hazard of foreign objects. Ergo, it's a CCP.

Whether you call one specific filter the CCP, or the filtration process itself (all 3) a CCP is up to you. But the decision you make on this issue will dictate how you set up your controls - i.e. what your process checks are to ensure your filtration is ok.

Jon

Dear Charles

Tut tut tut - A missile no

A cheque from ISO for continuous promotion of ISO 22000 maybe

All I am asking people is forget about the terminology, forget about 22000 ask yourself if using the decision tree (Codex) filtration would be a CCP or not.

Kind regards,

Tony

[Charles.C]

Dear Jon,

Thks for yr input. You seem to believe that HACCP is totally user-flexible regarding options for interpretation and implementation. And perhaps you are right.

To recapitulate the situation, AFAIK there are three, in-line, series filters which are all equally capable of 100 pct success for removing the hazard (unknown material and size characteristics).

As far as I can see, if any of the 3 filter is regarded as likely to randomly fail in a non-negligible frequency (but not too often!) and thereby creates a significant health risk, use of the standard Codex tree generates 3 possible CCPs via the well-known question as to whether the step is specifically designed to handle the hazard.

added - I had posted an alternative solution here but my interpretation incorrect. Sorry people.

It is possibly safer to avoid decision trees altogether and stick to a less "subjective" decision procedure such as risk analysis ??

Rgds / Charles.C





-

[Jon5]

Charles:

Thank you for your thoughts as well. I'm very much enjoying the forums the more I get involved and start to read. Here are a few more of my thoughts.

Regarding use of the decision tree, I provide a relevant quote below from the Codex. IN my own words, the tree is intended for use as a tool in making a decision regarding CCPs. It is not absolute, but helps you apply logic to your decisions as to whether a step should be a CCP or not. At the end of the day, your HACCP plan should (1) identify what the potential hazards to the consumer are, and (2) should provide adequate controls to eliminate those risks, or reduce them to an acceptable level, based on sound scientific rationale (that's the validation piece of your HACCP plan). If you can't provide adequate proof of due diligence in protecting the consumer, you're in a scary place legally. From the Codex:

<snip>

7. Determine Critical Control Points

(SEE PRINCIPLE 2)

3

There may be more than one CCP at which control is applied to address the same hazard. The determination of a CCP in the HACCP system can be facilitated by the application of a decision tree

(e.g., Diagram 2), which indicates a logic reasoning approach. Application of a decision tree should be flexible, given whether the operation is for production, slaughter, processing, storage, distribution or other. It should be used for guidance when determining CCPs. This example of a decision tree may notbe applicable to all situations. Other approaches may be used. Training in the application of the decision tree is recommended.

If a hazard has been identified at a step where control is necessary for safety, and no control measureexists at that step, or any other, then the product or process should be modified at that step, or at anyearlier or later stage, to include a control measure.

"

<snip>

Dear Jon,

Thks for yr input. You seem to believe that HACCP is totally user-flexible regarding options for interpretation and implementation. And perhaps you are right.

To recapitulate the situation, AFAIK there are three, in-line, series filters which are all equally capable of 100 pct success for removing the hazard (unknown material and size characteristics).

As far as I can see, if any of the 3 filter is regarded as likely to randomly fail in a non-negligible frequency (but not too often!) and thereby creates a significant health risk, use of the standard Codex tree generates 3 possible CCPs via the well-known question as to whether the step is specifically designed to handle the hazard.

added - I had posted an alternative solution here but my interpretation incorrect. Sorry people.

It is possibly safer to avoid decision trees altogether and stick to a less "subjective" decision procedure such as risk analysis ??

Rgds / Charles.C





-