41 replies to this topic

[Lyon.Lin]

For the revision of ISO 22000, I made a decision-making process as the attached pic. 

 

I hope someone can give me the advice to make it better for use.

 

[Lyon.Lin]

For the revision of ISO 22000, I made a decision-making process as the attached pic. 

 

I hope someone can give me the advice to make it better for use.

 

Sorry for lacked the attached pic

Attached Files

•  Decision tree for OPRPs and HACCP plan.pdf   40.74KB   1354 downloads

[Charles.C]

Sorry for lacked the attached pic

 

Hi Lyon,

 

from iso22000 -

 

OPRP
control measure (3.8) or combination of control measures applied to prevent or reduce a significant food safety hazard (3.40) to an acceptable level (3.1), and where action criterion (3.2) and measurement (3.26) or observation enable effective control of the process (3.36) and/or product (3.37)

 

 

No idea what 4M1E is but yr tree appears to not comply with the above basic requirement.

[012117]

Hi, Lyon.

 

If correct the 4M refers to Machine, Man, Method and Material while E is environment? 

 

the decision must satisfy the criteria of clause 8.5.2.4.1 Selection and categorization of control measure. The decision tree must be patterned from that one and it will include criteria as: is able to reduce to acceptable level, where the control measure located, if you are able to establish quantifiable limit and how fast can you react on the deviation.

 

4M1E more or less will be fitted to PRP.

[Lyon.Lin]

Hi Lyon,

 

from iso22000 -

 

 

No idea what 4M1E is but yr tree appears to not comply with the above basic requirement

Sorry, I have to change it from 4M1E to 3M1E, it's mean Man, Machine, Material, and Environment.

 

And when the decision tree needs to be used, that means the hazard which goes into the decision tree process are the significant food safety hazards. Whatever it is the OPRPs or HACCP plan, "prevent or reduce a significant food safety hazard (3.40) to an acceptable level "   is appearing in the definition of OPRPs and CCP, it won't be a question for me to write into the decision-making process.

 

So, love if you can specify what part of this tree needs to be improved or reinforce??

[Lyon.Lin]

Hi, Lyon.

 

If correct the 4M refers to Machine, Man, Method and Material while E is environment? 

 

the decision must satisfy the criteria of clause 8.5.2.4.1 Selection and categorization of control measure. The decision tree must be patterned from that one and it will include criteria as: is able to reduce to acceptable level, where the control measure located, if you are able to establish quantifiable limit and how fast can you react on the deviation.

 

4M1E more or less will be fitted to PRP.

Sorry I have to change the 4M1E into 3M1E.

And clause 8.5.2.4.1 said:  "For each of the control measures selected, there shall be an assessment of the following:"

 

The risk assessment for the control measures is after the decision-making process. And more importantly, will clause 8.5.2.4.1 be a reference for frequency and methods (clause 8.5.4.3)? 

 

Clause 8.5.4.3

For each OPRP, the monitoring method and frequency shall be proportionate to the likelihood of failure and the severity of consequences.

 

[Charles.C]

Hi Lyon,
 

 

So, love if you can specify what part of this tree needs to be improved or reinforce??

 

 

for starters -

 

(1) see the definitions

(2) compare yr tree -

 

 oprp.png   25.4KB   93 downloads

[Lyon.Lin]

Hi Lyon,
 

 

for starters -

 

(1) see the definitions

(2) compare yr tree -

 

oprp.png

 

OK, I see.

 

The route of it which u specify purpose to prevent the hazard from raw material, ingredients, packaging material, processing aids.

 

So the question I ask just focus on the eliminating or reducing.

 

For the OPRPs, it's only one way to go, prevention.

 

Thanks, I need to think about the other fittable question to fix that defect.

Edited by Lyon.Lin, 23 January 2019 - 08:12 AM.

[Charles.C]

OK, I see.

 

The route of it which u specify purpose to prevent the hazard from raw material, ingredients, packaging material, processing aids.

 

So the question I ask just focus on the eliminating or reducing.

 

For the OPRPs, it's only one way to go, prevention.

 

Thanks, I need to think about the other fittable question to fix that defect.

 

Actually there are already many (food) trees in use for (the previous) iso22000-2005 which are posted on this forum. Some  are probably still OK for food purposes with (perhaps) minor adjustments.

(Pharmaceutical requirements are maybe not exactly equivalent).

 

iso22000-2005 has (IMO) never had a satisfactory (food) methodology on how to differentiate CCPs/OPRPs. The latest version is attempting to remove some of the ambiguities.

[Lyon.Lin]

Yes, it is. So the decision tree shall be design based on the differences of the factory characteristic, not just follow one mode and do it forever.

 

In actually, when the ISO 22000:2018 release, I advocated myself not to use the decision tree, and how to make a decision for choosing the OPRPs or HACCP.

 

But now I figure out, the method for choosing control measures which in my brain is the decision-making process, and this process just based on the risk.

 

So I make it real, it is not possible for me to make a decision without this.

 

But it still has to present the variety in this decision tree for use of different factory, in actuality.

 

I think some significant hazards can be control by the OPRPs or HACCP plan (e.g. cross-contamination). And it's just base on the resources (money, time or etc...) what the organization have. 

 

So I still have to make a little change for this decision-making process.

[Charles.C]

Hi lyon,

 

Yes, some non-tree methods have been previously proposed here also (and in the Literature). One example is utilised in the model excel analysis here -

 

http://www.ifsqn.com...ge-7#entry50651

[Lyon.Lin]

Hi lyon,

 

Yes, some non-tree methods have been previously proposed here also (and in the Literature). One example is utilised in the model excel analysis here -

 

http://www.ifsqn.com...ge-7#entry50651

 

Don't you think the description "Control Measure Assessment Data (ISO 22000 Paras. 7.4.4 [a-g]) (Scoring  Low/Med/High = 1/2/3)" for OPRPs and HACCP plan choosing very unstable?

 

For No. 1C, the result of its score is 19, what if the food safety team changes or the responsibility for hazard assessment personnel change, then this result will go to OPRP (<18) ?

 

It just put too many risks on the personnel for decision making.

 

The risk of significant or non-significant can be verified by quantitive analysis, but for decision making of OPRPs or HACCP? I don't think it needs a range of risk. Because it is a risk itself. @@

 

Complexity for decision-making has to depend on the size of the organization, I think.

Edited by Lyon.Lin, 23 January 2019 - 09:20 AM.

[Charles.C]

All the methodologies involved in HACCP-related topics have subjective components.

 

I personally think that introducing the concept of OPRP was  unnecessary and over-complicated an already non-simple situation.

 

iso22004 remarks that the relative number of  CCPs and OPRPs  is not particularly  important as long as adequate control of a significant hazard can be achieved.

[Lyon.Lin]

All the methodologies involved in HACCP-related topics have subjective components.

 

I personally think that introducing the concept of OPRP was  unnecessary and over-complicated an already non-simple situation.

 

iso22004 remarks that the relative number of  CCPs and OPRPs  is not particularly  important as long as adequate control of a significant hazard can be achieved.

 

FYI

 

0.3.3.3 Hazard analysis – Operational process

 

Decisions taken in the application of HACCP should be based on science, free from bias and documented. The documentation should include any key assumptions in the decision-making process.

 

Maybe there's a way for decision-making without thinking process, But I think I know what is the difference between OPRPs and HACCP plan, then I can materializing it as a tree or put every tree step into risk range.

 

All due respect, what is the important things for implementation? Clarify what the OPRPs and CCPs are or just use the simple way to do hazard control? In your opinion, why the ISO even have to clarify their two terms rather than simple one term?

 

Not on purpose for provoking a dispute, just continual improvement for myself.

 

Sorry, if I make a indiscreet remark.

[Charles.C]

FYI

 

0.3.3.3 Hazard analysis – Operational process

 

Decisions taken in the application of HACCP should be based on science, free from bias and documented. The documentation should include any key assumptions in the decision-making process.

 

Maybe there's a way for decision-making without thinking process, But I think I know what is the difference between OPRPs and HACCP plan, then I can materializing it as a tree or put every tree step into risk range.

 

All due respect, what is the important things for implementation? Clarify what the OPRPs and CCPs are or just use the simple way to do hazard control? In your opinion, why the ISO even have to clarify their two terms rather than simple one term?

 

Not on purpose for provoking a dispute, just continual improvement for myself.

 

Sorry, if I make a indiscreet remark.

 

Hi Lyon,

 

No problem. The diffficulty in discussing is that iso22000 has an extremely confused history. Notably initially (and still) regarding OPRP.

 

The original reasons why ISO introduced OPRP were/are attempted to be clarified/detailed in iso22004.

 

For example -
.

 

ISO 22000 reorganizes the traditional concept of dividing control measures into two groups [prerequisites and
measures applied at critical control points (CCPs)] in a logical order for the development, implementation and
control of the food safety management system.

 

Control measures are grouped into three groups, as follows:

a) prerequisite  programmes  (PRPs)  that  manage  the  basic  conditions  and  activities;  the  PRPs  are  not
selected  for  the  purpose  of  controlling  specific  identified  hazards  but  for  the  purpose  of  maintaining  a
hygienic production, processing and/or handling environment (see 7.2 of ISO 22000:2005);

b) operational  prerequisite  programmes  (operational  PRPs)  that  manage  those  control  measures  that  the
hazard analysis identifies as necessary to control identified hazards to acceptable levels, and which are not
otherwise managed by the HACCP plan;

c) a  HACCP  plan  to  manage  those  control  measures  that  the  hazard  analysis  identifies  as  necessary  to
control identified hazards to acceptable levels, and which are applied at critical control points (CCPs)

 

 

Can also see this (ca 2005) "analysis" of iso22000's haccp content -

 

 haccp logic in iso22000.pdf   698.91KB   623 downloads

 

(note that iso22004:2005 and above attachment were issued prior to the introduction of fssc22000)

(there is a near Encyclopedia in the French Language ca 2005 et seq arguing about how to interpret iso22000's OPRPs. Unfortunately, afaik, the only authoritative  books related to iso22000's intended haccp logic are also in French.The immediate (2005) furore over the meaning of "OPRP" was one reason for the almost immediate issuance of iso22004 which attempted to clarify iso22000's haccp requirements. ISO22004:22005 does help but did not much resolve the quantitative haccp aspects, ie distinguishing OPRP/CCPs.)

 

You can see a large number of "OPRP"  threads on this Forum starting around 2005 if you wish to see the history. However a lot of it simply illustrates the confusion.

 

Sadly, IMO, the topic of OPRP remains confused. It is interesting to note the changes in definitions in the latest iso22000 which have removed some of the more "mystical" elements in the first version. But .......

[Mathieu Colmant]

Hi everyone,

 

Sorry to disturb you in this topic :)

 

The main issue between CCP and oPRP becomes clearer in ISO22000:2018.

But I know it is more like a boy-scout game...

So, CCP(3.11) has a critical limit (3.12). And the note 1 says that when you are on the wrong side of the limit, the product must be considered as potentially unsafe (sorry, I don't have EN version, so I retranslate from French. exact words in EN version can differ).

Let's find those potentially unsafe products... this is managed in 8.9.4. And there is a small tip in 8.9.4.2, release of products: If the critical limit was not met, you can't release the product, they must be managed following 8.9.4.3. So, let's go there :

Products that can't be released (so this includes all products not meeting critical limits) must be

a) reprocessed, so the hazard is taken below the limit

b) used in another way, if this has no impact on food safety

c) destroyed

 

So, a critical limit is black or white. For example, when a product is rejected by the metal detector (CCP), you can't sell it. You can destroy it (try to find the metal source before ) or you can "reprocess" it, for example, remove the contaminated slice from the package then push it through the metal detector again.

But, for temperature, this is less clear. If your limit is 4°C, do you throw every product reaching 4.1°C ? No, you will try to find what happened, check the curve and the duration of the problem, then decide if the product is good or not. So, this isn't critical, this is an action criterion and you have an oPRP.

 

If you want to make a decision tree, or shett or whatever, you should take into account the bullets in 8.5.2.4. I join my decision tree, but I need to check it in details with the new version.

 

Concerning the bias, those are difficult to avoid. You will need to make choices (for example I take that scientific abstract and not that one, maybe because I don't know it even exist...), so you are adding bias. The only thing we can try is to keep them as low as reasonably possible.

Attached Files

•  Decision Tree ISO22000-2018 EN.pdf   136.47KB   782 downloads

Edited by VeilleAlim, 28 January 2019 - 10:41 AM.

[Charles.C]

Hi everyone,

 

Sorry to disturb you in this topic :)

 

The main issue between CCP and oPRP becomes clearer in ISO22000:2018.

But I know it is more like a boy-scout game...

So, CCP(3.11) has a critical limit (3.12). And the note 1 says that when you are on the wrong side of the limit, the product must be considered as potentially unsafe (sorry, I don't have EN version, so I retranslate from French. exact words in EN version can differ).

Let's find those potentially unsafe products... this is managed in 8.9.4. And there is a small tip in 8.9.4.2, release of products: If the critical limit was not met, you can't release the product, they must be managed following 8.9.4.3. So, let's go there :

Products that can't be released (so this includes all products not meeting critical limits) must be

a) reprocessed, so the hazard is taken below the limit

b) used in another way, if this has no impact on food safety

c) destroyed

 

So, a critical limit is black or white. For example, when a product is rejected by the metal detector (CCP), you can't sell it. You can destroy it (try to find the metal source before ) or you can "reprocess" it, for example, remove the contaminated slice from the package then push it through the metal detector again.

But, for temperature, this is less clear. If your limit is 4°C, do you throw every product reaching 4.1°C ? No, you will try to find what happened, check the curve and the duration of the problem, then decide if the product is good or not. So, this isn't critical, this is an action criterion and you have an oPRP.

 

If you want to make a decision tree, or shett or whatever, you should take into account the bullets in 8.5.2.4. I join my decision tree, but I need to check it in details with the new version.

 

Concerning the bias, those are difficult to avoid. You will need to make choices (for example I take that scientific abstract and not that one, maybe because I don't know it even exist...), so you are adding bias. The only thing we can try is to keep them as low as reasonably possible.

 

Hi VeilleAlim,

 

Thks for input.

 

The translation may be partly to blame but, offhand, yr attachment is almost the most complicated decision tree i have seen so far. (But see PS below)

 

The word "impact" as first introduced in iso22000(2005) has, afaik, resisted any satisfactory explanation since that time. It seems to be equated to synergistic in the attachment.

 

 IIIRC, the utilisation of the concept of safe/unsafe product as a decision parameter was also invoked for the previous iso22000 ? Perhaps in the French Literature ? Afnor ?

 

Personally, I am hoping that iso22004 (2014) (FWIW) is accepted as having foreknowledge of the probable content in iso22000(2018) so remains valid.

 

Regardless, appreciate yr attachment.

 

PS - after looking more closely at the latest iso22000's section(s) on distinguishing OPRPs/CCPs, the requirements seem to be even more complicated than the previous version (!!!) so I can perhaps understand why yr decision tree has expanded accordingly. 

 

Maybe a good reason for continuing to avoid using iso22000.

Edited by Charles.C, 28 January 2019 - 12:36 PM.
emended

[Vinish Vijayan]

Would be really helpful if you guys could check and give feedback. I have found similar one for the old version and thought to give a try for with the new requirement. I hope i have covered all the points.

Attached Files

•  Assessment control Measures as CCP or OPRP.pdf   96.83KB   732 downloads

Edited by Food for life, 24 February 2019 - 11:49 AM.

[Lyon.Lin]

Would be really helpful if you guys could check and give feedback. I have found similar one for the old version and thought to give a try for with the new requirement. I hope i have covered all the points.

 

What if there was a significant hazard has been assessed to be controlled at CCP, but the organization lack the fund to buy the equipment for this CCP, how is the decision-making process working?

[Vinish Vijayan]

What if there was a significant hazard has been assessed to be controlled at CCP, but the organization lack the fund to buy the equipment for this CCP, how is the decision-making process working?

 

 

I thought of addressing this issue with this question with g, h and i. if the organization lacks fund to purchase then we will give less weightage to make it controlled by OPRP. 

[Charles.C]

In respect to locating significant hazards, iso22000:2018, (8.5.2.3) has followed FSMA in changing their text, ie  -

 

 the likelihood of its occurrence in the end product prior to application of control measures

 

I have yet to find a literature justification for this innovation. In fact, I have so far not even seen a meaningful (to me) explanation of the objective.

 

As noted in post 18, the attachment looks like an expansion of the methods posted here 10 years ago for 2005 version.

 

There are some apparent difficulties, eg (g)1 as stated would seem to automatically imply zero chance of a CCP/OPRP, by definition. Similar conceptual/quantitative difficulties exist for most of the other factors in (a-i).

 

I have myself previously used an analogous table for the iso22000:2005  version but, in retrospect, I thinks it's probably simpler to construct a decision tree since there is less opportunity for debate (not that the existing decision trees are all satisfactory, far from it IMO).

[Charles.C]

Would be really helpful if you guys could check and give feedback. I have found similar one for the old version and thought to give a try for with the new requirement. I hope i have covered all the points.

 

Hi FFL,

 

JFI I attach a modified version of yr Table with a simple example included.

 

 Categorizing CCP-OPRP, iso22000,2018,Ver 2.0.xlsx   12.33KB   1364 downloads

[Vinish Vijayan]

When you mentioned regarding a decision tree i was just curious and thought to give a try. love to get the feedback 

Attached Files

•  High.pdf   89.33KB   939 downloads

[Charles.C]

When you mentioned regarding a decision tree i was just curious and thought to give a try. love to get the feedback 

 

Hi FFL,

 

Thks for yr efforts.

The tree probably needs more detail.

Boxes/Flow 2,4 are also questionable IMO.

 

This published tree (Coca-Cola) developed for iso22000:2005 is still almost OK  IMO except perhaps (i) needing a response to 4.1b3 to be inserted somewhere, (ii) IMO changing the "necessary" to "feasible" in boxes 4/5) (iii) adding/inserting a box for OPRP/action criteria text adjacent to box 4.

(afaik the absence of specific responses to factors 4.1a/b0/b3/b4 [already contained in 2005 Standard] never prevented it from being an auditorially acceptable decision  tree).

 

 COCA COLA MICHIGAN STATE CCP-OPRP tree.pdf   96.37KB   954 downloads

[Lyon.Lin]

In accordance with the requirement from ISO 22000:2018, there's no hazard can be defined that control by the OPRP or HACCP plan only.

 

For example, the significant chemical hazard (e.g heavy metal or pesticide residue) from the raw material, it can be controlled by oPRPs (COA or report from the supplier) or CCP(timely detect by external inspection and go next step of the process until getting the result).

 

Therefore, all we need to know just only the difference between OPRP and HACCP plan.

 

And the clause 8.5.2.4.1, I don't think it is a base for decision tree making, the only thing we need to do which based on the clause 8.5.2.4.1 an assessment, not try to make a combination with the decision tree.

 

Base on the ISO 22000:2018 including clause 3.2, 3.30, 3.12, 3.11, 8.5.4.2, 8.5.4.3, 8.9.2.2, 8.9.2.3, 8.9.4.2

 

OPRP

1. The OPRPs can be used at the step or out of the process.

2. The action criterion only using for distinguishing the met criterion or not.

3. The action criterion only helps to contribute to the assurance that the acceptable level is not exceeded.

4. The monitoring method and frequency shall be proportionate to the likelihood of the failure and the severity of consequences.

5. When action criteria are not met, need to do further investigation whether the product is nonconformity.

 

CCP

1. The CCP is a step.

2. The critical limit is exceeded or not met, the product(semi or end product) become the potential product.

3. The critical limit shall ensure that the acceptable level is not exceeded.

4. The monitoring method and frequency shall be capable of timely detection of any failure.

5. When the product affected by a failure to remain within critical limits at the CCP, the product is nonconformity. (totally different with 2.)

6. The monitoring system shall include all scheduled measurements relative to the critical limit(s). Which is mean a HACCP plan may include multiple monitoring system schedules.