25 replies to this topic

[GMO]

After the 'zero' CCP thread which has outlasted most people's expectations, it got me wondering. What are the plants with 10 or more CCPs?

I think in a ready meal plant from memory we had the following (and there's an argument this is too many):

Oven cooking
Retort cooking
Steam cooking
Sauce pan cooking
Cooling
Metal detection (some lines)
X-ray detection (other lines)
Chilled storage (perhaps debatable as a CCP)

What else could you have and why? Obviously things are specific to your plant but I think it would be a really interesting debate.

[Charles.C]

Dear GMO,

Have just been looking at a summary of one for yoghurt (2011) with 8 CCPs and 4 OPRPs. Does that count ?

Rgds / Charles.C

[Madam A. D-tor]

Dear GMO,

I think that I do not have a client with more then 10 CCPs.
But I can give a list with the most frequent found CCPs:
1) temperature of incoming materials. Ofcourse only for dairy, meat, fish, chilled products. Even though some producers have a heating step, authorities here require this is a CCP. The say: because it is a legal requirement. I personally disagree that a legal requirement is automatically a CCP, but that is how it is here.
2) temperature control of storages;
3) heating processes;
4) cooling down after heating;
5) O₂-concentration for MAP packing;
6) foreign body detector;
7) temperature during transport.

Discutabele:
8) no leakages for automatic vacuum pack line.

sometimes found:
9) pH for acid products;
10) adding additives/preservatives. (too low concentrates will lead to spoiled or not safe food and too much might be chemical contamination)
11) coding (if the BBD is coded wrong, consumers will eat unsafe products)

I believe that organisations with more then 7 CCPs, also have more HACCP-plans/product groups.
In your example you name 8 CCPs. However 1 product will never be involved in all CCPs. E.g. A product will undergo metal detector OR X-ray, and probably the product will undergo oven cooking OR retort cooking OR steam cooking. ( I wonder, was this product pasteurised in the packaging or heat filled? Why are these no CCPs? Ultra fresh meals with very short shelf life?)
So in one HACCP-plan related to one product, I do not think there can be more then 5 CCPs, but I am just like you very interested.

[mesophile]

Hi All,

I remember auditing a factory whilst working for a UK retailer, the company had 26 CCP's, one of which was stress of staff!

I asked, how can stress be a CCP? The answer was, if the staff are stressed they may make a fundimental food safety mistake resulting in non-conforming product leaving the site. As staff are the last people to see the product before it leaves the building, there are no further steps that could rectify the issue.

I then asked, how do you control this CCP? The answer was "by giving the staff 20 days holiday per year, and paying them 10p/hour above minimum wage". At first I thought they were joking, however they were deadly serious.

It just shows how a little bit of knowledge can be a dangerous thing!

Regards

Simon

[Charles Chew]

Hi All,

I remember auditing a factory whilst working for a UK retailer, the company had 26 CCP's, one of which was stress of staff!

I asked, how can stress be a CCP? The answer was, if the staff are stressed they may make a fundimental food safety mistake resulting in non-conforming product leaving the site. As staff are the last people to see the product before it leaves the building, there are no further steps that could rectify the issue.

I then asked, how do you control this CCP? The answer was "by giving the staff 20 days holiday per year, and paying them 10p/hour above minimum wage". At first I thought they were joking, however they were deadly serious.

It just shows how a little bit of knowledge can be a dangerous thing!

Regards

Simon

Hi Simon, I really like this one. Imagine what could happen if they run the regulatory. I can fully understand the impact of "mental hazard" but I am dead keen to know the verification frequency on mental stress and the CL validation of this CCP. Truly scary stuff!

[onsolution]

It just takes a bit of imagination.

How about cameras throughout monitoring faces? Then you can measure Smile Per Hour. If it gets a bit low, then off for a break. Of course there are those people who can't smile to save themselves, but you probably don't want them in a stress free workplace.

Or go for a slightly faster method - get everyone together to watch "The Office". The you just have to gauge the amount and volume of laughter.

Loud laughter - fine
Soft laughter - fine, just introverted
Shaking of head, tears - highly stressed, work related
Calling out "hey, that's Pete" - serious problems

The extra pay is a nice touch, but they should have considered the relationship between "downsizing" and the deteriorating quality of food.

[GMO]

Dear GMO,

I think that I do not have a client with more then 10 CCPs.
But I can give a list with the most frequent found CCPs:
1) temperature of incoming materials. Ofcourse only for dairy, meat, fish, chilled products. Even though some producers have a heating step, authorities here require this is a CCP. The say: because it is a legal requirement. I personally disagree that a legal requirement is automatically a CCP, but that is how it is here.
2) temperature control of storages;
3) heating processes;
4) cooling down after heating;
5) O₂-concentration for MAP packing;
6) foreign body detector;
7) temperature during transport.

Discutabele:
8) no leakages for automatic vacuum pack line.

sometimes found:
9) pH for acid products;
10) adding additives/preservatives. (too low concentrates will lead to spoiled or not safe food and too much might be chemical contamination)
11) coding (if the BBD is coded wrong, consumers will eat unsafe products)

I believe that organisations with more then 7 CCPs, also have more HACCP-plans/product groups.
In your example you name 8 CCPs. However 1 product will never be involved in all CCPs. E.g. A product will undergo metal detector OR X-ray, and probably the product will undergo oven cooking OR retort cooking OR steam cooking. ( I wonder, was this product pasteurised in the packaging or heat filled? Why are these no CCPs? Ultra fresh meals with very short shelf life?)
So in one HACCP-plan related to one product, I do not think there can be more then 5 CCPs, but I am just like you very interested.

Interesting examples. I also disagree with No 1. If "legal requirement" is a CCP, is the checkweigher a CCP?! I think it can be a CCP but it depends on your process.

I'm interested in number 5. I would argue again this is more likely to be a quality issue than a food safety one. What's likely to grow if O2 % is OOS? Is it actually dangerous? It probably depends on the product.

I'd be fascinated how you monitor number 8!

Surely number 10, spoiled food is a quality issue, and is too much preservative really a food safety one?

For coding, perhaps it's a language issue but if you're considering BBD, then it's definitely not a food safety issue because where dates are critical, they are use by dates (and even then it's arguable whether thats a CCP. Most chilled food I've worked in with short shelf life spoils long before there's a food safety issue so I'd argue it would be apparent to the consumer!)

[Charles Chew]

Dear GMO,

I think that I do not have a client with more then 10 CCPs.
But I can give a list with the most frequent found CCPs:
1) temperature of incoming materials. Ofcourse only for dairy, meat, fish, chilled products. Even though some producers have a heating step, authorities here require this is a CCP. The say: because it is a legal requirement. I personally disagree that a legal requirement is automatically a CCP, but that is how it is here.
2) temperature control of storages;
3) heating processes;
4) cooling down after heating;
5) O₂-concentration for MAP packing;
6) foreign body detector;
7) temperature during transport.

Discutabele:
8) no leakages for automatic vacuum pack line.

sometimes found:
9) pH for acid products;
10) adding additives/preservatives. (too low concentrates will lead to spoiled or not safe food and too much might be chemical contamination)
11) coding (if the BBD is coded wrong, consumers will eat unsafe products)

I believe that organisations with more then 7 CCPs, also have more HACCP-plans/product groups.
In your example you name 8 CCPs. However 1 product will never be involved in all CCPs. E.g. A product will undergo metal detector OR X-ray, and probably the product will undergo oven cooking OR retort cooking OR steam cooking. ( I wonder, was this product pasteurised in the packaging or heat filled? Why are these no CCPs? Ultra fresh meals with very short shelf life?)
So in one HACCP-plan related to one product, I do not think there can be more then 5 CCPs, but I am just like you very interested.

I would like to know about the V & V on no.8 while I think Coding control under no. (11) should really be supported by "Pre-Start Up" and "Positive release Management". I am in agreement with all the rest.

[gcse-fhp]

quote name='GMO' timestamp='1326185224' post='51002']
After the 'zero' CCP thread which has outlasted most people's expectations, it got me wondering. What are the plants with 10 or more CCPs?
[/quote]

Thank you for your post. It leaves me curious about "zero" CCP. Let's say an operation prepares meat products for sale to further processing by other operations or consumers. There are grinding, formulation, blending, pre-cooking/cooling, pastry preparation, filling and packaging steps in this operation.

Opening of ingredient metal cans is also done. There is no foreign material detection at any point in the process.

All of the products are sold frozen. Does such an operation need to establish any CCPs, or is this a "zero" CCP operation?

Thanks for your input.

gcse-fhp

Edited by gcse-fhp, 22 January 2012 - 09:20 PM.

[Madam A. D-tor]

I'm interested in number 5. I would argue again this is more likely to be a quality issue than a food safety one. What's likely to grow if O2 % is OOS? Is it actually dangerous? It probably depends on the product.

Yes, O2 concentrate depends on product. For red meat O2 is mostly added to maintain the red colour. --> No CCP.
For chilled RTE O2 is lowered to reduce growth of aerobic bacteria including pathogens e.g. Listeria and Salmonella. It also has to be higher then 0,2% to reduce the growth of Clostridium B. and other anaerobic pathogens.There are scientifical reports for this from TNO-Netherlands. However I do not have these. Limits for these kind of products are mostly 0,2%-1% O2. If only 1% is used, I like to see a validation why anaerobic pathogens are no issue.

I'd be fascinated how you monitor number 8!

I would like to know about the V & V on no.8

I do not know what you mean with V & V, but I assume it is like monitoring, critical limits, corrective actions.
I have seen several monitoring methods, which are all also known from sealing checks on sealed packings.
1) after vacuum machine the packing (in boxes) is done manual. Workers have to touch the product and will notice when the packing is not vacuum. They need to record the failed packs in an hour. The NC pack is repacked and vacumised again. If there are more then X NCs in an hour the supervisor need to be informed, line will be stopped and vacuum machine will be checked.
2) every hour one packing for each mold is visually checked in the seal. Record if seal is intact/not intact. If not packs from the last hour need to be visualy inspected and if needed repacked.
3) every hour one empty packing is made and with a needle piched and with a manual pump, air is pumped in the packing. If 50 pumps do not lead to damaged packing it is good.
4) every hour a empty packing is made and is cut open by the operator. The thickness and strongness of the seal is checked.

ad 1) I had one customer with this CCP and it lead to discussions every visit. Also because the critical limit was set on 50 leakages an hour and they would find a maximum of 6 an hour. (but normally no leakages were observed). It was more a process control monitoring. They have, after 6 years, let go of this CCP. They pack fresh red meat, used for cutting up and further processing by butchers. These customers have enough knowledge to know the difference between vacuum and not vacuum and will sent back the products if this is wrong.

A similar CCP/monitoring is seal checks on sealed packings. Methods 1, 2 and 4 seen for these checks, but also checks with a water basin (if there are no visual bubbles, there is no leakage), with pressure tank or with specific leakage detector equipment. However most common for sealed packings is method 1. This CCP was quite common for 5-6 years ago, but is more and more reduced to 'just' a quality issue or a process monitoring.

Surely number 10, spoiled food is a quality issue, and is too much preservative really a food safety one?

Depends on the wording of the hazard in the risk assessment. Are we sure these are not carcinogenic? If the hazard is documented as a double dose it is a different risk then if the hazard is documented as a little bit more.
But I agree, that I do not see this CCP often anymore.
In feed industry adding additives is almost always a CCP.

For coding, perhaps it's a language issue but if you're considering BBD, then it's definitely not a food safety issue because where dates are critical, they are use by dates (and even then it's arguable whether thats a CCP. Most chilled food I've worked in with short shelf life spoils long before there's a food safety issue so I'd argue it would be apparent to the consumer!)

Yes, you are right. I meant used by date (I always forget thesre are also 2 different codes in English language). I think these are rather critical in RTE. There are surely ready to eat products where pathogens can grow, before spoilage is visible. Consumers are not very smart relating to food products/ food safety. We have discussed in another thread, that the majority of the consumers throw away products on the BBD/ CBD/UBD even if the product is still fine (e.g. whole fresh fruit, milk, yoghurt etc) My parents in law were eating cream cake and were saying that it had a very funny taste, but just eat it anyway because the label said that it would still last for 3 days and it wat bought that moring. When I tasted it, I recognised the cream to be spoiled. As you know this is very good recognisable and even though they tasted that it was nog good, they rely on the coding and eat it anyway. Ready to eat meals are mostly foreign tastes. I mean it is mostly: Chinese, Thai, Indonesian, indian, Mexican, Italian, Greek, Cadjun, etc. All of them with strong and sometimes unfamilair tastes. I do not know If I would recognise an off flavour or off taste in these products. So when there was a mistake in the code printer say, instead of 12-01-2012 it was 11-02-2012, I am sure some consumers will eat it and might get sick.

I think Coding control under no. (11) should really be supported by "Pre-Start Up" and "Positive release Management".

Of course this is a start up thing, but why can a start up check not also be a CCP?
If it is documented as a part of the positive release, then it surely is a CCP.

Edited by Madam A. D-tor, 23 January 2012 - 12:20 PM.

[George @ Safefood 360°]

From experience over the years conducting food safety audits, working in industry and providing services to clients I have come up with a new series of CCP decision questions...

1. Is your regulatory inspector insisting a specific step is a CCP? Yes - It's a CCP No - It's not a CCP
2. Is your most important customer insisting a specific step is a CCP? Yes - It's a CCP No - It's not a CCP
3. Are you fearful that a specific step might be a CCP but not really sure? Yes - It's a CCP No - It's not a CCP

and so on...

I'm not attempting to be cynical about a very important issue for food businesses but very often the decision making process is driven more by these factors rather than a full and critical assessment of the hazard and risk.

I strongly believe that the CAC CCP decision tree has contributed to the confusion regarding CCP's and needs serious review. Bottom line - a step is either a CCP or it's not! And where there is doubt or strong influencing opinions at play more information needs to be obtained.

I mentioned to Charles C in a recent post that I conducted some work on developing an alternative model (decision tree) for determining CCP's. I will be posting this in the coming weeks and really look forward to some community feedback.

[Charles Chew]

I do not know what you mean with V & V, but I assume it is like monitoring, critical limits, corrective actions.
I have seen several monitoring methods, which are all also known from sealing checks on sealed packings.

Sorry for the delay in replying. V & V stands for validation and verification.

[GMO]

Of course this is a start up thing, but why can a start up check not also be a CCP?
If it is documented as a part of the positive release, then it surely is a CCP.

I disagree. Start up checks are normally not CCPs because by nature they are infrequent monitoring. A CCP check would normally require frequent monitoring (e.g. minimum hourly or ideally continuous). I also disagree that anything as part of a positive release programme is a CCP. For example, in chocolate manufacturing, part of my positive release was the date code. Definitely not a CCP but more because there had been instances in that factory of 'humourous' disaffected staff members changing the date code to swear words etc.

[Madam A. D-tor]

I disagree. Start up checks are normally not CCPs because by nature they are infrequent monitoring. A CCP check would normally require frequent monitoring (e.g. minimum hourly or ideally continuous).

I have thought about your words, but keeps to my question: why can a start up check not also be a CCP? A start up check will be performed by every product start up. Depending on the product this can be from weekly to every ten minutes. If the coding is identified by an organisation as CCP and the equipment is validated not to change by itself, this can be enough. Normally a coding is checked more often.

I also disagree that anything as part of a positive release programme is a CCP. For example, in chocolate manufacturing, part of my positive release was the date code. Definitely not a CCP but more because there had been instances in that factory of 'humourous' disaffected staff members changing the date code to swear words etc.

Well, maybe you are right. I have never been in a company with positive release procedure, but the way I always thought of it: "if a check is so important that you do not sell your products before the results of that check are verified, it must be really important." Probably because I am food safety focussed I changed that in "it is really important to food safety".

[Charles.C]

Dear All,

Quoting Madam A.D-tor -

I have thought about your words, but keeps to my question: why can a start up check not also be a CCP? A start up check will be performed by every product start up.

Presumably, the nearest solution to an equitable result will be to do a risk assessment, accepting its subjectivity. At least one can then argue over semi-quantifiable matters.

Although one will still be faced with deciding whether prerequisites can be used to control "high" risks or not.

Rgds / Charles.C

[GMO]

Yep but there are "really important" checks which aren't CCPs. In the confectionery factory where proper positive release was in place on the basis of monitoring (not micro), the staff also were asked to attach a sample of the packaging (with code) and outer case label. This also ensured the correct packaging went on. There could have been allergen implications of them not getting this right but tbh, the errors were more likely to be with promotional packaging used in error or similar. Not a CCP but still very important to the business and a big cost if you get it wrong.

We also had a taste panel as part of the positive release process and again, not a CCP.

I see what you mean though that if you validate something doesn't change once checked it could be a CCP to have a once a shift check. I was also not thinking about high frequency change over products. Even then though, for CCPs I generally demand a start and end of run check (e.g. metal detection). I would also argue that the CCP check frequency must be related to the product release. A flour mill I know of for example, only checks their sieve once a week because it's impractical to check more often. That sieve is a CCP. Now they release product on an hourly basis. If that sieve failed, which is possible but doesn't happen often, then not only will the flour have been delivered to customers, it will have been used and in the shops.

[Charles.C]

...... Bottom line - a step is either a CCP or it's not! ...........

Dear George,

Codex approved ( ) but a probability tag might be revealing. One could argue that the CCP decision is irrelevant and attention is more usefully focused on the step's "event health significance / choice of control" aspects. Unfortunately i dislike ISO 22000's associated modus operandi even more intensely than Codex.

I look forward to seeing yr innovations.

Rgds / Charles.C

[Charles.C]

Dear GMO,

Interesting example.

That sieve is a CCP. Now they release product on an hourly basis. If that sieve failed, which is possible but doesn't happen often, then not only will the flour have been delivered to customers, it will have been used and in the shops.

I find it difficult to believe that this set-up can be considered an acceptable implementation of haccp. Unless all concerned actually accept that the step is not a real CCP but only a GH type 1-3.

So what do the flour company / receivers do when there is a failure?

Rgds / Charles.C

[GMO]

Dear GMO,

Interesting example.



I find it difficult to believe that this set-up can be considered an acceptable implementation of haccp. Unless all concerned actually accept that the step is not a real CCP but only a GH type 1-3.

So what do the flour company / receivers do when there is a failure?

Rgds / Charles.C

They say they would recall. I don't think it's acceptable either. Presumably they must replace the sieve before it fails? Still not something I'm comfortable with. I'd love to see their BRC audit report!

[Charles Chew]

A flour mill I know of for example, only checks their sieve once a week because it's impractical to check more often. That sieve is a CCP. Now they release product on an hourly basis. If that sieve failed, which is possible but doesn't happen often, then not only will the flour have been delivered to customers, it will have been used and in the shops.

It is common for flour mills to have a sieving step before final packing and this step is certainly a CCP. However, the sieve design often consist of multi-tier sieves within the same housing and as such not pragmatic for daily inspection. But there is no cause for concern as the mesh size of multi-tiered sieves are homogeneous so if one sieve is compromised, the others will take care of the problem.

[Madam A. D-tor]

It is common for flour mills to have a sieving step before final packing and this step is certainly a CCP. However, the sieve design often consist of multi-tier sieves within the same housing and as such not pragmatic for daily inspection. But there is no cause for concern as the mesh size of multi-tiered sieves are homogeneous so if one sieve is compromised, the others will take care of the problem.

So, which sieve is then the CCP? The first? The last? Or all together?

In a company working with powders, I found once the following CCP (However IMO it could have also been a PRP ) It wasan automatic line for dosing or something.
CCP = sieving
monitoring =check if sieve is in place/check if sieve is placed by the operator, before production
method =: visual
record = sieve is or is not placed in line;
critical limit = sieve is not in place
corrective action: = place sieve in line before start of production.

I do not think this is a good CCP, but it came just in mindwhen reading above posts.

[Charles Chew]

So, which sieve is then the CCP? The first? The last? Or all together?

The design of the sieve is under one housing with multiple sieve trays within (maybe about 10 removable sieves). The question you raised is valid under normal circumstance. However, because the porosity sieve size is the same for each tray within the same housing, the design is built to ensure flour size homogeneity as well as a measure to sieve out foreign particles. The housing for the sieiving is the CCP. Unless you have all sieves damaged, it is unlikely to occur but the entire sieve housing is indeed the last step for removal on a continuous process basis as a single step.

[Charles Chew]

In a company working with powders, I found once the following CCP (However IMO it could have also been a PRP ) It wasan automatic line for dosing or something.
CCP = sieving
monitoring =check if sieve is in place/check if sieve is placed by the operator, before production
method =: visual
record = sieve is or is not placed in line;
critical limit = sieve is not in place
corrective action: = place sieve in line before start of production.
I do not think this is a good CCP, but it came just in mindwhen reading above posts.

critical limit = sieve is not in place - What is the sieve porosity size that determines the "kill step" for trapping

[Charles.C]

Dear All,

An interesting discussion about sieves but i see no mention of risk assessment, eg what is the likelihood of the sieve/housing failure in this specific case ? Has there ever been an actual significant health consequence in this specific case ? Or any case for this product in a comparable process/step ?

It seems to me that the CCP conclusion reached implies, very crudely, that LxM = CCP ? If so, the rest of the hazard analysis should look quite interesting. For example, as notionally introduced in previous post, are "fine" filters being solely used to protect against "boulders" or already well-filtered product?

I reference the USA semi-standard concept - Only hazards which are reasonably likely to occur are a candidate for significance. IMHO, this can be "reasonably" interpreted as M to H but excludes L. Unfortunately it also necessitates an, at least, semi-objective evaluation of the actual case as against just seizing one extreme ( CCP) or the other (PRP).

Looks like another example of GH-CCP 1-3 to me.

Rgds / Charles.C

[GMO]

Dear All,

An interesting discussion about sieves but i see no mention of risk assessment, eg what is the likelihood of the sieve/housing failure in this specific case ? Has there ever been an actual significant health consequence in this specific case ? Or any case for this product in a comparable process/step ?

It seems to me that the CCP conclusion reached implies, very crudely, that LxM = CCP ? If so, the rest of the hazard analysis should look quite interesting. For example, as notionally introduced in previous post, are "fine" filters being solely used to protect against "boulders" or already well-filtered product?

I reference the USA semi-standard concept - Only hazards which are reasonably likely to occur are a candidate for significance. IMHO, this can be "reasonably" interpreted as M to H but excludes L. Unfortunately it also necessitates an, at least, semi-objective evaluation of the actual case as against just seizing one extreme ( CCP) or the other (PRP).

Looks like another example of GH-CCP 1-3 to me.

Rgds / Charles.C

Yep, flour mills often have considerable levels of grain contamination on incoming wheat, you should see the **** that comes in . Whether the last sieve is the CCP or not is a question worth asking but they often don't have metal or x-ray detection systems either so the only controls are magnets (not ideal) or sieves so at least one of those sieves will be a CCP.

Also I've known of two instances of foreign body contamination of flour which would suggest the sieves aren't always perfect. Depending on the sieve type as well, if the overtailings are not frequently checked (i.e. what's taken out), you can get the contamination feeding back into the flour as well as not being aware that there is an issue earlier in the process so I think it is a CCP IMHO.