Jump to content

  • Quick Navigation
Photo

RMS/ MRL - sample size for multi-site

Share this

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

509fruit

    Grade - Active

  • IFSQN Active
  • 4 posts
  • 0 thanks
0
Neutral

  • United States
    United States

Posted 22 July 2025 - 05:31 PM

Hello, Long time lurker. USA

Residue Monitoring System- MRL

 

Currently switching from Global Gap option 2 to SQF v9 primary production modules 2, 7 and IPM addendum. Multi-Site.

 

It will be a super group after two entities merge into the new entity (whole other fun time). Overhauling everything right now.

 

Looking at how entity A (already SQF certified but will release its cert for the merge)  did samples for their smaller operation doesn't work for us. While at the same time we feel the MRL sample size seem excessive with no legal exceedances noted in the history of either RMS program.

 

The residue monitoring system is one where there have been issues of interpretation on sample size in GG with auditors and CB.

SQF doesn't have any guidance when I have looked.

 

Previously it was set to the square root of each commodity of group members. For (not perfect) example 7 apple, 7 cherry and 7 pear. (both conv and organic sample lots taken)

 

Is there any real guidance on sample size?

Would providing the evidence of no incidences from the rms history's be allowed to factor in for a reduced sample size?

 

Note the growers have no decision on where their fruit goes to market. The warehouse tests the fruit, sales desk decides its MRLs are good for certain export markets.

 

 

side note:

It took forever to figure out FSMA requires just 1 water sample now, so orchards are set for 1 sample of water before harvest since SQF sides with legal regulation. Global gap required 3 and in the past FSMA was 4-5 samples.

 

Thank you in advance for your insight.

 


  • 0

Scampi

    Fellow

  • IFSQN Fellow
  • 6,113 posts
  • 1649 thanks
1,849
Excellent

  • Canada
    Canada
  • Gender:Not Telling

Posted 08 August 2025 - 12:15 PM

I would defer to AQL table to determine the sample size that makes good sense and based in science


  • 0

Please stop referring to me as Sir/sirs


509fruit

    Grade - Active

  • IFSQN Active
  • 4 posts
  • 0 thanks
0
Neutral

  • United States
    United States

Posted 08 August 2025 - 04:02 PM

That is a nice tool.  The one thing is that even if there are some residues detected but not exceeded, or one market is hits hot (ex India with their influx limits now at 0.01 for everything). The sales desk just finds the appropriate home for the fruit. There is a history of 0 true rejects due to residue levels. Farmers spray to the EPA labels on the chemicals. Fruit always tests fine for domestic.

 

I went here to try AQL Table | ANSI AQL Chart | AQL Calculator | AQI Service

For fun lets use a number of 85.  I would then on a table say order quanity 85,  inspection level 1 for lowest.  not entirely sure I set AQL at 6.5 as that seemed lowest, I cannot list all the countries a sample is screened against so 6.5 seems close to the top 10 markets list. 

How do I make sense of the information below?

The table spits out this: Sample Size: 8.  Accept Point: Reject Point: 2


  • 0

Scampi

    Fellow

  • IFSQN Fellow
  • 6,113 posts
  • 1649 thanks
1,849
Excellent

  • Canada
    Canada
  • Gender:Not Telling

Posted 08 August 2025 - 04:05 PM

That means if you have 1 sample that is OUTSIDE the  allowable range, the lot is ACCEPTABLE   

 

2 or more, the lot would be a reject


  • 0

Please stop referring to me as Sir/sirs


509fruit

    Grade - Active

  • IFSQN Active
  • 4 posts
  • 0 thanks
0
Neutral

  • United States
    United States

Posted 08 August 2025 - 04:30 PM

I'm not sure how to have this apply in tree fruit for outside and acceptable.

   It sounds like the sample from 12 growers out of say 85 as a group number of growers, and of org and conv among cherry pear and apple may work as the sample size is larger than the AQL table. And a sample of 8 would not cover the spread of commodities and org/conv.

 

again there has never been a failed mrl test (where the fruit is exceeded for export or domestic market), it just wont be shipped to that country if exceeded so no rejected growers.

the warehouse receives copies of spray records before receiving fruit so we know what is sprayed on the incoming fruit.


  • 0

Scampi

    Fellow

  • IFSQN Fellow
  • 6,113 posts
  • 1649 thanks
1,849
Excellent

  • Canada
    Canada
  • Gender:Not Telling

Posted 08 August 2025 - 05:42 PM

I was suggesting it as a starting point

 

In my history, each LOT required testing---------so the AQL would be based on the lot size

 

So, you receive say, 10 trailers of apples (I know each trailer may have 30 totes) but use 10 as the lot size, then AQL will tell you how many samples to pull from that lot, so maybe it's 50

so 50 apples become your COMPOSITE sample that is then tested

 

Yes, it's a lot

 

No SQF does not provide REGULATORY guidance

 

You can have "skip lots" whereby based on total annual volume you can skip say every 3rd lot (where a single commodity is a lot)

 

The multi site here should not be a factor---your sampling should be solely based on incoming lots (and if there is no requirement (as some organic sprays do not have a required MRL) you get to skip that lot)

 

For this particular item, I would NOT accept history as a reason to reduce testing


  • 0

Please stop referring to me as Sir/sirs


509fruit

    Grade - Active

  • IFSQN Active
  • 4 posts
  • 0 thanks
0
Neutral

  • United States
    United States

Posted 08 August 2025 - 06:52 PM

Last year when under global gap v5.4-1 we were told the square root of the group entity members, rounded up,.  last year the math and group size came to about 21-28 samples depending on the organic needs and sales desk requests. (~9apple, ~9 cherry, ~9pear).

 

Was trying to figure out if staying with the same sample program or less is possible due to the level of control we have on the fruit and the history of 15+ years of no issues.

 

in the tree fruit industry they do not test by bin lot # received for mrl testing. We cannot do composite or pooled fruit as that makes the sample inaccurate for traceability.

 

"No SQF does not provide REGULATORY guidance"  this was what I was trying to understand, where GG is prescriptive on things.


  • 0



Share this


Also tagged with one or more of these keywords: RMS, MRL, tree fruit

1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users