7.4.4 d talks about the likelihood of failure of control measure but not likelihood of occurence of hazard? My understanding is that if the likelihood of occurence of a hazard is high, then it is more likely to be controlled as CCP.
It is important to follow the sequence of paragraphs in iso22000.
7.4.3 deals primarily with the determination of significant hazards. The decision as to whether a CM is CCP/OPRP remains “open”.
7.4.4 deals with “proposing/evaluating/validating” control measures (CM) which can “handle” the determined significant hazards. And the categorization of a validatable CM option to CCP/OPRP.
However, if the control measure is easy to fail, then the hazard will not be securely controlled, we should select another control measure which is stable so that the hazard can be securely controlled. I wonder if this is a selection criterion rather than categorization criterion? I am still stuck at this concept.
iso22004 (2005) suggests (not demands) a degree of prioritization (for a CCP) towards elements (abe) within (a-g). Such an approach typifies Tree methods (eg Procert) although in practice the choice of prioritization varies between authors. In contrast, Modarres’s method uses equal weighting to obtain an 'average", other variations of Modarres use unequal weights. (Note that iso22004 (2014) may have now added further suggestions, have no idea.)
But in practice most auditors, so far, typically seem to not care regarding subtleties such as the above. They simply require a “logical” method for section 7.4 which is “relatable” to the iso standard. Note that iso22004 (2005) comments that it does not really matter if you conclude CCP or OPRP provided that FS is achieved (ie CM validation regarding achieving the acceptable level exists) and monitoring is feasible within an adequate time frame.
For synergistic effects, may I use your yogurt hazard analysis as example. In the "Raw milk inspection and off loading into Raw Milk Silo" step, the control for chemical is a CCP. The score for 7.4.4 (g) was 3, would you kindly help to use this example to explain more on synergistic effects? e.g. what are the pair/group of synergistic control measures relevant to this control measure?
TBH, I should have slightly updated the above “CCP”. As per the excel comments, I struggled conceptually regarding this specific CM. I would now tend to classify it as a PRP (via ISO22002-1) since the result is less likely to be auditor debated.
This is an idea of Synergy –
A practical illustration of Synergy is here –
I wouldn’t worry too much about about this element unless it’s a well-documented situation.
Regarding the excel yoghurt analysis “(g)”, note that I modified Modarres’s scoring concept, (see the note in sheet 4).
The above text illustrates what is IMO a major defect in the iso22000 standard. Section 7.4.4 is highly non-prescriptive other than that the output must be either CCP, OPRP, or go back to square 1.
IMO the optimum procedure is probably one which (a) suits yr process, (b) takes the least analysis time, © minimizes yr subsequent workload, (d) satisfies the auditor. Not necessarily in that order.
Rgds /Charles.CGo to the full post