9 replies to this topic

[Rancy Sharma]

Hi Everyone,

 

 

I have a question about FSMA HACCP. We are Canadian company and have strong BRC HACCP established. Is it possible to combine FSMA- HACCP and BRC- HACCP together or we need to make it separately?

 

Any information will be greatly appreciated!

 

Thank you

[Charles.C]

Hi Everyone,

 

 

I have a question about FSMA HACCP. We are Canadian company and have strong BRC HACCP established. Is it possible to combine FSMA- HACCP and BRC- HACCP together or we need to make it separately?

 

Any information will be greatly appreciated!

 

Thank you

 

Hi Rancy,

 

I haven't checked but I thought there was a version of Food BRC in which the "haccp" is supposed to be "FSMA compatible".

[Ryan M.]

Yes.  Just add a column in your BRC Hazard Assessment for "Preventive Control Measures for Human Food (FSMA) Required (Process, Allergen, Sanitation, Supply Chain)".

 

Then you will have a "Preventive Controls Summary" like you have a HACCP Hazard Summary.

 

See attached screenshot of a line from our hazard analysis.

Attached Files

•  Capture.PNG   218.48KB   2 downloads

[Nikki R]

Yes it is possible. I was signed up to emails a little while ago for a lady called Kassy Mash with techni -K who wrote a series on how to comply to both standards at once.

This is the link to the newsletter archive, hope it helps

https://techni-k.co.uk/FSMA-Articles

[Charles.C]

Yes.  Just add a column in your BRC Hazard Assessment for "Preventive Control Measures for Human Food (FSMA) Required (Process, Allergen, Sanitation, Supply Chain)".

 

Then you will have a "Preventive Controls Summary" like you have a HACCP Hazard Summary.

 

See attached screenshot of a line from our hazard analysis.

 

Hi Ryan,

 

Thks for the detailed screenshot although I'm a bit puzzled at the risk assessment scheme/entries/calculation.

 

Is "1" the maximum level of severity ? (Presumably not Low?)

Is "C" a medium level of Likelihood of Occurrence ? (seemingly concluded as Low)

 

How do you end with "4" ? (an additive scheme although 25 implies not ?)(non-linear scales ?) and what is a "significant" result ?

 

I also (a lot) wondered what the last (chopped) column was titled/stated ?  YES ?....[to ?] 

 

Regardless this looks like an elegantly logical/simple add-on appproach to FSMA if it's auditorially OK. Many thanks for sharing.

[Ryan M.]

Hi Ryan,

 

Thks for the detailed screenshot although I'm a bit puzzled at the risk assessment scheme/entries/calculation.

 

Is "1" the maximum level of severity ? (Presumably not Low?)

Is "C" a medium level of Likelihood of Occurrence ? (seemingly concluded as Low)

 

How do you end with "4" ? (an additive scheme although 25 implies not ?)(non-linear scales ?) and what is a "significant" result ?

 

I also (a lot) wondered what the last (chopped) column was titled/stated ?  YES ?....[to ?] 

 

Regardless this looks like an elegantly logical/simple add-on appproach to FSMA if it's auditorially OK. Many thanks for sharing.

 

That was only half of the picture Charles.  The first sheet in our hazard analysis workbook deals with the measuring of the hazard significance and frequency of occurrence.  This is assessed by the ingredients & materials related to each process step, personnel practices, environmental hazards (within the facility), recall history, equipment risk as it relates to a specific hazard type, etc.  See attached.

Attached Files

•  Capture2.JPG   100.43KB   0 downloads

[Charles.C]

That was only half of the picture Charles.  The first sheet in our hazard analysis workbook deals with the measuring of the hazard significance and frequency of occurrence.  This is assessed by the ingredients & materials related to each process step, personnel practices, environmental hazards (within the facility), recall history, equipment risk as it relates to a specific hazard type, etc.  See attached.

 

Hi Ryan,

 

Thanks for pic. I now recall this is a published SQF matrix with non-linear, asymmetric scales.

 

If I understand correctly the conclusion in column L of 1st pic is that the step analysed is (risk = 4) significant despite being controlled by GMP.

 

Just for contrast, I enclose another published matrix with similar qualitative logics but linear, symmetrical scales. The interpretation for analogous inputs is now non-significant.

 

Illustrates the subjectivity of risk assessment.

Attached Files

•  risk matrix.png   166KB   0 downloads

[Ryan M.]

Hi Ryan,

 

Thanks for pic. I now recall this is a published SQF matrix with non-linear, asymmetric scales.

 

If I understand correctly the conclusion in column L of 1st pic is that the step analysed is (risk = 4) significant despite being controlled by GMP.

 

Just for contrast, I enclose another published matrix with similar qualitative logics but linear, symmetrical scales. The interpretation for analogous inputs is now non-significant.

 

Illustrates the subjectivity of risk assessment.

 

Yes, quite subjective.  The key to the risk assessment and hazard analysis is to backup any claim or decision.  

[FurFarmandFork]

I actually rewrote my own matrix to make it more accessible to my food safety team, as well as provide a clear line between hazards requiring a PC or not (I hate the matricies that use a "yellow" section), and allow room for the "likelihood" section to take into account FDA hazards guidance, which may otherwise fall off my scale for actual likelihood but are expected to be seen in my plan.

 

 

Attached Files

•  Austin's likelihoodxseverity table.pdf   54.1KB   132 downloads

[Charles.C]

I actually rewrote my own matrix to make it more accessible to my food safety team, as well as provide a clear line between hazards requiring a PC or not (I hate the matricies that use a "yellow" section), and allow room for the "likelihood" section to take into account FDA hazards guidance, which may otherwise fall off my scale for actual likelihood but are expected to be seen in my plan.

 

Hi 3F,

 

Still contentious IMO due the notoriously ambiguous "Likelihood of occurrence in the absence of a preventive control". And the "10"s. But no doubt very FSMA-friendly.

 

Just by their nature/design risk matrices exhibit varying degrees of linearity/risk averseness. And likelihood/severity biasing.