Jump to content

291Online Users

  •  
- - - - -

Risk Assessment Matrix


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

#1 Guest_Jean_*

Guest_Jean_*
  • Guests
  • thanks

Posted 15 October 2007 - 06:07 AM

Hi!

Can anybody provide the risk assessment matrix for hazard analysis. Previously I had done using Likelihood (L,M,H) and Severity (L,M,H).

Thanks in advance,

J


  • 0

#2 Gaskit

Gaskit

    Grade - MIFSQN

  • IFSQN Member
  • 106 posts
  • 1 thanks
0
Neutral

  • Jamaica
    Jamaica
  • Gender:Male
  • Location:Greater Manchester

Posted 15 October 2007 - 08:09 AM

In preparation for version 3 of the BRC/IoP Packaging Standard where now we are required to produce a risk management matrix sec 1.9, I knocked this together, hope it can be of help (although I do not like utilising such matrix as borderline cases always get reduced to the lower category).

Regards,

Steve

New link added 10022008 to replace broken original / Charles.C -

Attached File  Risk_Matrix_steve_gasket_saferpak_.xls   18KB   985 downloads

(also available as No1 in Ch.C post in this thread, ca. 09022008)


  • 0
I know God will not give me anything I cann't handle, I just wish that he didn't trust me so much.

Thanked by 1 Member:

#3 Charles.C

Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 6,781 posts
  • 1392 thanks
104
Excellent

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

Posted 15 October 2007 - 03:38 PM

Dear Steve,

A nice classical example.

I have never had an auditor query me as to my method of classifying between for example occasional/seldom or marginal/negligible. They typically look at my slightly varied English explanations (has happened / may happen etc) for each and rapidly (hurriedly?) pass on to the final CCP’s. It depends on the specific case of course but this step is surely about as subjective as it gets. I won’t mention oPRPs.

How about you ? :smile:

Rgds / Charles.C


  • 0

#4 Madam A. D-tor

Madam A. D-tor

    Grade - SIFSQN

  • IFSQN Senior
  • 419 posts
  • 119 thanks
7
Neutral

  • Netherlands
    Netherlands
  • Gender:Female
  • Interests:meat, meat products, ready to eat, food safety, QMS, audits, hazard analyses, IFS, BRC, SQF, HACCP, ISO 9001, ISO 22000

Posted 15 October 2007 - 07:56 PM

Dear Steve,

I have never had an auditor query me as to my method of classifying between for example occasional/seldom or marginal/negligible. They typically look at my slightly varied English explanations (has happened / may happen etc) for each and rapidly (hurriedly?) pass on to the final CCP’s. It depends on the specific case of course but this step is surely about as subjective as it gets. I won’t mention oPRPs.




Hi Charles,

I really have to react on that!
It really depends on the systems you were audited against (is this proper English?). During BRC or IFS audits there is too less time to do an in-depth investigation of the HACCP analyses. When I am auditing the Dutch HACCP Requirements, I really evaluate the HACCP analyses. But, only once in three years or when changes or amendments are made.

I think mentioning the OPRPs and PRPs is a good thing. It makes it very clear that those risks are taken care of, that they are controlled in one way or another.

Don't worry! No defence! I have seen both sides of the fence myself and as an auditee, I always hated it, if the auditor did not take time to look at the important things. I'm trying to do a better job now, but as you probably know, it is not as easy as it seems.


Dear Jane and Steve,

There are also matrices with 5 X 5 or 3 X 3 occurrence and severity.


Steve,

With your matrix you are saying that a hazard which does not occurs very often, but has a catastrophic (death?) effect is no CCP.
Can you give me an example of this? What is the definition you used for unlikely? And what for seldom?
  • 0
Kind Regards,

Madam A. D-tor

#5 Guest_Jean_*

Guest_Jean_*
  • Guests
  • thanks

Posted 16 October 2007 - 08:04 AM

Dear All,

Thanks for your input.

Madam A.D-tor:</U> Can you advice as from where to obtain info on matrices with 5 X 5 or 3 X 3 occurrence and severity.

During my surveillance audit, the auditor had asked me to review Hazard Analysis using the risk assessment matrix which was a suggestion from his side.

I request you to provide some assistance in the risk assessment for below:-
Process: No cook step (catering operation)

Step: Preparation

Hazard : Biological (M, C) - due to improper handling,poor personal hygiene, poor cleaning &amp;&nbsp;&nbsp; sanitizing, temperature abuse

Risk assessment:- Likelihood –M ; Severity- H

Using the Codex decision tree I have decided to consider this step as CP.
But as per the auditor, this step should be a CCP because severity is High. This is my concern
My decision to consider this as CP was based because we rarely had any food complaints in past 1 year or half from such foods, we have good system in place to monitor staff personal hygiene and we verify the cleanliness of hands and food contact surfaces every month by microbiological analysis. Moreover the cold food temperatures are monitored and maintained below 7oC during storage with shelf life of 3 days. In buffet we refill every second hour or as and when required to ensure the display foods deviating from critical limit of 8oC are kept for only 4 hours and after which this is discarded. (Only little quantity is being displayed to control food costs.)

What do you folks recommend?

BR,

J


  • 0

#6 Gaskit

Gaskit

    Grade - MIFSQN

  • IFSQN Member
  • 106 posts
  • 1 thanks
0
Neutral

  • Jamaica
    Jamaica
  • Gender:Male
  • Location:Greater Manchester

Posted 16 October 2007 - 08:33 AM

Dear Madam A.D-tor,

Thanks for your comments, as I attempted to convey, utilising a matrix in my opinion is a waste of time as there is always the borderline cases, anyhow here goes with the explination:

UNLIKELY TO OCCUR: As opposed to cleaning packaging machinery with Food Area Degreaser, one of the operatives would utilise Non Food Degreaser and a dirty rag full of C.DIF and not rinse it of effectively, this if transfered to food stuff (RTE for example) in processing would in my opinion be catastophic, however; THIS IS UNLIKELY TO OCCUR due to all relevant OPRP and PRP in place.

SELDOM: Transport companies turn up with non food items on the same load as food grade items (external contractors utilised by suppliers on occasion), it seldom happens, but on occasion it has / may do in the future, again mainly controlled by OPRP / PRP, but one can not always control the mind set of a trucker.

A little far fetched, however I hope it provides a little food for thought.

Kind regards,

Steve


  • 0
I know God will not give me anything I cann't handle, I just wish that he didn't trust me so much.

#7 Charles.C

Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 6,781 posts
  • 1392 thanks
104
Excellent

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

Posted 16 October 2007 - 11:14 AM

Dear Madam ADT,

Yr query regarding high severity producing a low risk illustrates the difficulty of providing exact answers. Without looking it up in detail, I think it is common to see a prioritisation (subjective obviously) on the side of “likelihood of occurrence”. For example can have a look at pgs 17,18 in this reference where there is a variety of interpretations of the consequence/ likelihood parameters –

LINK BELOW HAS REPLACED ORIGINAL LINK WHICH WAS BROKEN (charles.c / 09022008)
Attached File  risk_matrix_chc.doc   480.5KB   767 downloads

Jean refers to a “buffet”, this is probably regarded as a high risk topic, (does it also include things like oysters ?) which will cause some preliminary sharpening of auditor knives in all likelihood (and I guess rightly so from an epidemiological viewpoint).

No mention is made regarding literature validation of any CCPs for the above – I would expect the IT to offer substantial help for such a sensitive process ? Presumably the auditor had a similar query ? Similarly a validated non-CCP would be equally meaningful, you can see some comments regarding hygiene defects in the above ref. but the actual process will presumably dominate. I have found that providing one can validate one’s interpretations, an auditor will tend to be relatively non-argumentative even if he/she disagrees with yr opinion ( providing that he/she does not have a superior counter-reference :smile: , this has rarely happened IMEX with well-documented systems but it depends on the process again).

One has to remember that probability is the basis of the above via historical / scientific data whatever so some people prefer numbers as in the worked example in this forum (dwl section) but you still have the same borderline / linear /exponential decisions to make. I would suggest that there is also somewhat of a temptation to choose the matrix format so as to achieve a self-perceived requirement (reverse probability). All these complexities promoted the Codex tree option but this still necessitates a risk analysis somewhere even though it’s often not explicitly shown, in this respect ISO22k should perhaps be congratulated. Or not ?

I didn’t initially notice Steve’s inclusion of the prp evaluations in his upload. This is the first example of a published matrix on this site which includes such AFAIK. Very generous IMO (and brave). :clap: :clap:

Rgds / Charles.C

added - Personally, I have only seen one other example of an ISO 22000 oriented risk matrix analysis on the IT, namely at this link - http://www.nordicinn..._img/heggum.pdf
This latter approach has, I think, preemptively removed the "prps" from consideration within the risk matrix stage so that only oprps and CCPs potentially remain, this is perhaps more closely aligned to the I22k philosophy ?


  • 0

#8 Bobby

Bobby

    Grade - AIFSQN

  • IFSQN Associate
  • 35 posts
  • 0 thanks
0
Neutral

  • United Arab Emirates
    United Arab Emirates
  • Location:Dubai
  • Interests:To Enjoy Making, Eating and Studying FOOD

Posted 16 October 2007 - 06:50 PM

"I request you to provide some assistance in the risk assessment for below:-
Process: No cook step (catering operation)

Step: Preparation

Hazard : Biological (M, C) - due to improper handling,poor personal hygiene, poor cleaning &amp;&nbsp;&nbsp; sanitizing, temperature abuse

Risk assessment:- Likelihood –M ; Severity- H

Using the Codex decision tree I have decided to consider this step as CP.
But as per the auditor, this step should be a CCP because severity is High. "


Jean,
What is the Critical control point and the critical limit the auditor is looking at in this step?????
Goes on to show that HACCP auditing is miserable when it comes to catering operations!
My professor used to say " If you want a perfect system, every step is a CCP!"

however, do not rely on the number of complaints to identify a CCP.

try answering the following questions:

Did the auditor find something that indicated poor personal hygiene?

Do the food handlers wear gloves when they handle ready to eat foods?

How about hand and food contact surface swab test results?


Bobby Krishna
Food Inspector
Dubai Municipality
  • 0

#9 Guest_Jean_*

Guest_Jean_*
  • Guests
  • thanks

Posted 17 October 2007 - 05:44 AM

Dear Bobby,

Thanks a lot for your response!



Did the auditor find something that indicated poor personal hygiene?

No

Do the food handlers wear gloves when they handle ready to eat foods?

Gloves are not mandatory in the prep. area as long as our staff wash hands correctly. and yes, some of our staff do use gloves during mixing etc. He didn't notice any poor working or hygiene practices from our staff during the operation.

How about hand and food contact surface swab test results?

hand swabs are checked both internally and externally from all the high risk operations staff every month including food contact surfaces in these areas and we had no swab failures in these areas.
The auditor was very happy with our facility.

To be frank he was suggesting, since the severity is High, the preparation step should be considered as a CCP. I had explained & shown him all the documents supporting my decision and he didnt make a non-conformance on that, instead suggested to use the risk assessment matrix and review the hazard analysis......this was for improvement.

BR,

J


  • 0

#10 Madam A. D-tor

Madam A. D-tor

    Grade - SIFSQN

  • IFSQN Senior
  • 419 posts
  • 119 thanks
7
Neutral

  • Netherlands
    Netherlands
  • Gender:Female
  • Interests:meat, meat products, ready to eat, food safety, QMS, audits, hazard analyses, IFS, BRC, SQF, HACCP, ISO 9001, ISO 22000

Posted 17 October 2007 - 09:21 AM

Madam A.D-tor:</U> Can you advice as from where to obtain info on matrices with 5 X 5 or 3 X 3 occurrence and severity.

Hi Jean,

I have no links or documents available at this moment. I will look this up this weekend and come back to you.

I request you to provide some assistance in the risk assessment for below:-
Process: No cook step (catering operation)

Step: Preparation

Hazard : Biological (M, C) - due to improper handling,poor personal hygiene, poor cleaning &amp;&nbsp;&nbsp; sanitizing, temperature abuse

Risk assessment:- Likelihood –M ; Severity- H


It depends on your used definition on high severity and medium Likelyhood. If for example high severity means death can occur and medium likelyhood is something between monthly and once in 3 months, your company management is actually saying: "We accept that once in 3 months someone dies from our products". Probably your company won't take that responsability. So you have to make sure things can not go wrong, and set a CCP. A CCP is the highest protection you can give to the customers despite from forbid them to eat your products.
It is really a theoretical problem. It may occur and the effect may be very high. So you have to take measurements. If you are saying by your risk assessment that people can die from it, and after that decide it is not important enough to control by a CCP, it makes a strange of your company. Probably in production every thing is controlled very well, because you do not seem to have any problems (complaints, bad results on hygienic checks, etc.)
The auditor should also report if you are able to perform a good HACCP analyse. This is an example that seems strange in theoretical way but is probably allright in the practical way. But maybe you make the same "mistake" with other hazards, which was the auditor not able to detect.
That is why (in my opinion) an auditor should always discuss the HACCP-analyses. To find out if the company understands what they are doing and if they are using an adequat method.

I hope this explenation (in poor English) helped you in oone way or another.
  • 0
Kind Regards,

Madam A. D-tor

#11 Charles.C

Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 6,781 posts
  • 1392 thanks
104
Excellent

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

Posted 17 October 2007 - 11:26 AM

Dear Jean,

I think you should be thanked for initiating this thread since some very interesting comments being made.

With all respect to the previous posts, I suggest that you detail the process a little more if some further specific advice required. I think this was the idea behind Bobby’s comment ?.


(BTW, the non-use of gloves for RTE items is a major defect in some Food Codes, USA for example, although previous threads here have shown very mixed opinions on this subject.)

Eg, What is the product? Chicken sandwiches ?

Eg, What is the Hazard ? It is usually demanded IMEX (by the auditor also) to specify a microbiological hazard, not generalise by (MC) as above, eg >> Salmonella. This demonstrates that the auditee understands the process and focuses the risk assessment. (Perhaps you have x-referenced this elsewhere in the HACCP plan but my auditors would not accept such a presentation).
Assuming Salmonella, I certainly agree with Madam DTA’s comment that a routine occurrence of Salmonella every 3 months in a RTE food would be unsatisfactory ! Perhaps more like catastrophic !

Risk assessment:- Likelihood –M ; Severity- H


I find yr own assessment of “M” a little surprising/worrying, why not "L"?, ie, how did you decide ? Does this reflect the specific input (cooked) materials you are using or your own assessment of the surrounding hygiene (the latter ok from yr previous post)

What risk matrix are you / the auditor using at the moment ?

There is a freely dwlable 3 x 3 matrix on this site thanks to Simon (> General Discussion>Doc Exchange>HACCP Manual Example for Packaging ). This uses numbers and you can see that M x H does not lead to a CCP although the auditor might not agree with this matrix of course. A comment could be that there is a problem with the matrix size limiting the options, especially for a sensitive product/process. If you compare with the matrix in Steve’s or my post, you can see a CCP is equally possible there (= a return to the subjective aspect). Nonetheless, if you are able to validate yr result to the auditor for an analogous process, this should help, can you? (I rather doubt the auditor wud accept the one here as a valid xref (no lit.ref. :rolleyes: ) but maybe ??)

Rgds / Charles.C

added - BTW - @Madam A. D-tor - Yr English is excellent and yr posts are very much appreciatd :thumbup:
  • 0

#12 Guest_Jean_*

Guest_Jean_*
  • Guests
  • thanks

Posted 18 October 2007 - 10:58 AM

Dear Charles / Madam D-tor ,

I am really thankful and glad for all your valuable inputs.

I agree with Charles that microbiological hazards should be specified, but in my case (catering industry) where HACCP is process based and have more than 2000 different menu and many with different methods of preparation and it would be too cumbersome to conduct a detailed HA for each menu in retail operations. I feel this can be better done in small restaurants.....

But we have categorized different products like Meat, seafood’s, poultry, diary products etc..and have specified the microbiological hazards like Salmonella in poultry instead of chicken sandwich, chicken curry etc....

Charles, anybody would be surprised / worried when they see risk assessment done with "M" and not "L" as long as the person does not work or know the operations of the particular facility.
My decisions were based on my observations during inspections, checking staff competency, operations (during different business seasons), kinds of consumers etc. which is again subjective.
Currently I am not using any risk matrix and have decided to work out in detail.

As Madam A.D-tor has mentioned risk assessment is theoretical. From my point I consider the severity "H" for vulnerable people who may consume foods likely to make them sick due to lack of awareness and finally blame the food processors / caterers. Moreover all the steps in the food flow are considered and have appropriate controls in place irrespective of whether it is a CCP or not.

I strongly agree that an auditor should assess hazard analysis and the responsible person should be able to give valid reasons and documents to support the decision.

I am already working on the review of HA to ensure no hazards have been missed and this thread has been very valuable.

Madam: Last of all, please I am not concerned about the English being used for communication all the comments from Steve, Bobby, Charles and yourself have been very much appreciated and benefited.

BR,

J


  • 0




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users