100% checking or random sampling?

Do you think a random sample of raw materials from the supplier would suffice in the process of accepting deliveries? I mean, how many data need to check, 100% or random sample?
Thank you in advance.

Dear MOM,

Are you checking chilli peppers by consumption ?

Sorry, couldn't resist.

I suggest you categorise yr situation to enable an answer of "random by how much."

Statistical textbooks often recommend a minimum 30 samples in case of total ignorance. The Central Limit theorem is usually rather less demanding.

Rgds / Charles.C

Dear MOM,

Are you checking chilli peppers by consumption ?

Sorry, couldn't resist.

I suggest you categorise yr situation to enable an answer of "random by how much."

Statistical textbooks often recommend a minimum 30 samples in case of total ignorance. The Central Limit theorem is usually rather less demanding.

Rgds / Charles.C

Ok, we should categorize the items/materials. However, in that case, we cannot make a policy statement something like “only acceptable materials from approved supplier shall be received. “

Do you think a random sample of raw materials from the supplier would suffice in the process of accepting deliveries? I mean, how many data need to check, 100% or random sample?
Thank you in advance.

Your sampling should be based on risk and historical performance. So what are the risks associated with the raw material and have there been any problems in the past?

Your sampling should be based on risk and historical performance. So what are the risks associated with the raw material and have there been any problems in the past?

Based from the responses so far it seems that the organization has some freedom in selecting the sample, but the problem is what might be considered a sound statistical representation. Tony-C said that it should be based on risk and historical performance. What I want to know now is, is the assessment of probability and impacts of hazard a requirement of ISO 22000 during receiving of deliveries from supplier? I mean this activity comes prior to activation of control measures (e.g. PRPs, OPRPs, and CCPs).

N+1 square root method can be considered.....
There are actually three versions of the rule:

1. Take the square root of the lot size to get the sample size. Accept on zero defects. This version of the rules determines the sampling plan to use.

2. Take the square root of the number of cartons, open that number of cartons and select the required number of samples from them. In this case the sampling plan including the number of samples is already determined. For example, if the sample size is 20 and 50 cartons exist, SQRT(50)+1=8 cartons must be opened. Select 3 samples from 4 cartons and 2 from the remaining 4. The square root of n plus one rule is used to obtain a representative sample.

Take the square root of the number of drums, sample from this number of drums and composite the samples together to run a single test. For example, if 50 drums are received, take samples from SQRT(50)+1=8 drums, composite them together and measure the characteristic of interest. The rule is used again to obtain a representative sample.

Regards
Jomy Abraham

Do you think a random sample of raw materials from the supplier would suffice in the process of accepting deliveries? I mean, how many data need to check, 100% or random sample?
Thank you in advance.

Dear MOM,

Ok, we should categorize the items/materials. However, in that case, we cannot make a policy statement something like “only acceptable materials from approved supplier shall be received. “

It depends what you are precisely responding to / what you are actually doing (???), but a (general) policy statement could be like “Raw materials are sourced from approved suppliers and further sampled for acceptability at time of reception”

The relevant (main)(QP) procedure(s) (see Tony’s post) is typically listed individually at end of section and/or cumulatively at end of manual, the chosen format being stated at the beginning of manual. Sub-procedures are then specified in the main procedures as I’m sure you know already.

It would probably be helpful if you posted a typical flow chart / specification for your product X. Otherwise, some of our answers may well be nonsense.

Sampling plans are invariably a compromise between objectives / capabilities / logistics. Any (your) chosen plan may determine the probability of hazards existing in yr starting material. Only you know the possible significance of such details. for example, see ISO 22004, par.7.3.

Rgds / Charles.C

It depends what you are precisely responding to / what you are actually doing (???), but a (general) policy statement could be like “Raw materials are sourced from approved suppliers and further sampled for acceptability at time of reception”

Better. I liked this statement. Thank your for your suggestion.

The relevant (main)(QP) procedure(s) (see Tony’s post) is typically listed individually at end of section and/or cumulatively at end of manual, the chosen format being stated at the beginning of manual. Sub-procedures are then specified in the main procedures as I’m sure you know already.

I think so.

It would probably be helpful if you posted a typical flow chart / specification for your product X. Otherwise, some of our answers may well be nonsense.

The documents that I am developing are for our subsidiary located in another site. I haven’t been there but was scheduled to visit them next week to prepare them for ISO 22000 certification comes this May 2011. I will their flowchart here as soon as I saw there processes. I was tasked to document our subsidiary’s FSMS manual. My boss said they are certified to HACCP already so I presume and very hopeful that they already established PRPs, OPRPs and CCPs. Otherwise, it would be a disaster since I lack knowledge of it.

Sampling plans are invariably a compromise between objectives / capabilities / logistics. Any (your) chosen plan may determine the probability of hazards existing in yr starting material. Only you know the possible significance of such details. for example, see ISO 22004, par.7.3.

Actually I am not trying to structure it along the lines of ISO 9001. It’s just that I was used to it as once a user of ISO 22000 myself. On the other hand, I have no experience on ISO 22000.

Its is mainly based on your FG requirement. If you are concentrating in Mycotoxins/Pesticides, you have to do a good sampling - Refer EC178/2010.
In general - also refer ISO 2859 for deciding your sampling plan.

Do you think a random sample of raw materials from the supplier would suffice in the process of accepting deliveries? I mean, how many data need to check, 100% or random sample?
Thank you in advance.

Dear MOM,

Thks for yr comments.

From the previous post, I guess I must hv missed something in yr other posts and yr product X is related to either groundnuts (peanuts), other oilseeds, tree nuts, apricot kernels, liquorice or vegetable oil ?

My boss said they are certified to HACCP already so I presume and very hopeful that they already established PRPs, OPRPs and CCPs.

I hope you realise that (Non-ISO 22000) HACCP has never utilised OPRP (although you might be slightly lucky if yr site has specialised in Dutch HACCP).

It’s just that I was used to it as once a user of ISO 22000 myself. On the other hand, I have no experience on ISO 22000

.

Sorry but I didn’t understand the above.

Rgds / Charles.C

Mind Over Matter, do you want to come back on the comments made by Charles C?

Dear MOM,

Thks for yr comments.

From the previous post, I guess I must hv missed something in yr other posts and yr product X is related to either groundnuts (peanuts), other oilseeds, tree nuts, apricot kernels, liquorice or vegetable oil ?

Fresh fruits and vegetables
Ready-To-Drink (Bottled juice)
Ready-To-Drink (Canned juice)
Frozen fruits
Preserved fruits (e.g. mangoes, etc)
Dried sweetened fruits

I hope you realise that (Non-ISO 22000) HACCP has never utilised OPRP (although you might be slightly lucky if yr site has specialised in Dutch HACCP).

If that is the case, i'm not sure if they have OPRP.
.

Sorry but I didn’t understand the above.

Rgds / Charles.C

Sorry if my statement was not clear.

Thanks for your comment Charles. Thank you Simon as well.

N+1 square root method can be considered.....
There are actually three versions of the rule:

1. Take the square root of the lot size to get the sample size. Accept on zero defects. This version of the rules determines the sampling plan to use.

2. Take the square root of the number of cartons, open that number of cartons and select the required number of samples from them. In this case the sampling plan including the number of samples is already determined. For example, if the sample size is 20 and 50 cartons exist, SQRT(50)+1=8 cartons must be opened. Select 3 samples from 4 cartons and 2 from the remaining 4. The square root of n plus one rule is used to obtain a representative sample.

Take the square root of the number of drums, sample from this number of drums and composite the samples together to run a single test. For example, if 50 drums are received, take samples from SQRT(50)+1=8 drums, composite them together and measure the characteristic of interest. The rule is used again to obtain a representative sample.

Regards
Jomy Abraham

However, if the underlying distribution deviates from normal distribution to a highly skewed distribution, then the accuracy of the 95% confidence statement with sample size obtained from the square root of N plus one rule decreases