5 replies to this topic

[frankiet]

Hello - just wondered if anyone would be able to assist me with the following?

Just to set the scene - recently an error occurred in the factory where I work, when we accidently packed off some product after check weighing it at the incorrect weight (by this I mean that the incorrect test weights were used to calibrate the machine - these were 20g less than the test weights that should have been used)

No one realised until we had approximately 1.5 days worth of production - now we're left with product...or potential product that may fall below the T2 weight. We're now faced with having to re-weigh this - but I just wanted to know what options are available to us. As it stands the individual units that travel over the check weigher, have been packed into sleeves (each) - and into an outer case.

I guess what I'm trying to find out is whether we're able to weigh them in the sleeve - and do this manually - as opposed to bringing them back into a production areas to re-run them down the production line. What are the rules here - i.e. packing to average weight manually and not having the use of an in-line check weigher. Also - wanted to be reminded of the ratio of checking should we decide to go down this route.

Hope this all makes sense - thanks to anyone for your advice

[Foodworker]

Sounds like you have a big problem!

You don't say how many units are involved, but what you now have to prove is that the batch complies with the 3 packers rules.

The most valid way of doing this is to weigh every unit individually. This can be done manually or by running them over a checkweigher. Only you can decide on what is practical in your factory.

For either route, you should be able to recalculate the tare weight of the unit's packaging and sleeve . The exception to this is if there is a significant variability in the packaging weight which would invalidate using a standard tare. (From memory I think the sd has got to be less than 0.1 x the TNE, but don't quote me)

If individual weighing is not an option there is the possibility of batch blending to rectify the error, but the limitation here is that you may have individual packs below the T2 which you must remove.

There are some published guidelines, which unfortunately I can't lay my hands on - called something like The Code of Practical Guidance for Packers, although it is practical in the sense that you need a PhD in statistics. If I recall, it is very cheap to buy, TSO bookshops should have it.

The other alternative is to be upfront with your local Trading Standards and ask for their help. They often like to called upon rather than be viewed as only enforcers.

[Charles.C]

Dear frankiet,

You might find the attachment in this post (and thread) of interest (esp. if you hv a PhD )-

http://www.ifsqn.com...dpost__p__36561

And also this thread may be of interest -

http://www.ifsqn.com...dpost__p__33053

Rgds / Charles.C

[frankiet]

Thanks Foodmaker and Charles for responding - some good advice there! We'll have to get it right first time in future.

Thanks again

Frankiet

[GMO]

The packers guide is out of print unfortunately. We got a sneaky photocopy from our TSO.

Now weight control is something which is poorly understood by pretty much everyone (sorry to say this but even by some TSOs!) I've seen blatently illegal packs on sale before due to the positioning of the weight declaration but that's another issue.

There is nothing wrong with a manual system; however, you've not said what your target weights are but they are unlikely to be 20g above nominal I would think? There is a chance you could use a manual sampling system but I can tell you now that the likely outcome is that your average weight will be below nominal which is illegal. Unless I suppose the packs are filled by a filler which was not altered from its normal settings? (Although if that's the case, how disappointing that no-one noticed!)

As said before, there is nothing wrong with weighing in packaging. From memory, you should weigh 20 individual sets of packaging and if the packaging tare standard deviation of this group is < 1/10 TNE then it's ok to treat it as a constant. (Although presumably you should have done this for whatever packaging it's in already even if the sleeve isn't there.)

Here are some guidance notes;

http://www.nmo.bis.g...august 2007.pdf

Also once you do get everything resolved, it's a good idea to set up a dummy T2 pack that gets passed across your checkweigher every shift. That would have picked up this issue and limited the problem.

Unfortunately it's been an age since I've used a manual system and I suspect you will have to aim for a higher average and T1 / T2 settings to account for sampling error, however, my copy of the packers guide is at work (I'm on maternity leave, sorry). As I said, I suspect you will not be in a position to release this product even with using a manual system as the manual sampling is likely to show the product is illegal. Your quickest result would probably be to reweigh across a checkweigher but resetting your packaging tare to include the sleeve (if it doesn't vary too much).

Oh and I predict now that some bright operations spark will suggest releasing the product anyway "because it's not dangerous". I urge you to say "no". Although it's not a food safety issue, it is a legal issue and if you let your "grey area" extend that far, where will it stop...?

[Foodworker]

Just looked at the TSO Bookshop website and the Packers Guide is available at £9.50