Sample size calculation for liquids

Six Sigma – iSixSigma Forums Old Forums General Sample size calculation for liquids

Viewing 8 posts - 1 through 8 (of 8 total)
• Author
Posts
• #38464

Fil
Participant

I am trying to detect the presence of a microorganism in a liquid.  Right now, sample of 5mL is taken from various tanks ranging from 4 liters to 120 liters.  This sample undergoes bioburden test to detect any growth. How do I know what is the approppriate sample size for this?

0
#115063

Tim F
Member

I think there are two different questions lurking here, and they have two very different meanings and two very different answers. I’ll give my \$0.02 and see who has any better ideas.QUESTION 1) Is 5 mL an appropriate size sample to draw?
This questions depends on the test being done. If the test can be adaquately performed on 5 mL and consistently detect contamination at the levels you are interested in, then that is really all you need to know. The appropriate volume to draw for each test is determined by the test. QUESTION 2) How many samples should be drawn? This one is tough – tougher than deciding how many widgets to pull from a lot from sampling. You see, standard sampling assumes that the defects are randomly distributed. I’m assuming that by the time that a region is contaminated, a 5 mL sample will contain thousands or millions of nasty microorganism. If — and this is a big “if”) — if the contamination is uniformly distributed, then that presumably mean that ANY 5 mL sample will contain thousands (give or take a few hundred) of microorganisms. If, however, the microorganisms are concentrated in one area — maybe they start growing in one region and only slowly spread to other parts of the liquid — then you need to take enough samples to have a good chance of finding the bad region. That requires a knowledge of how the micro-organisms are actually distributed and that can only be determined by more careful study.
Tim

0
#115067

Szentannai
Member

Hi,
I’d ask how sensitive the measurement method is – i.e. what is the smallest concetration of the microorganism that you can detect assuming uniform distruibution. It would also be good to know whether you can homogenize the tanks befor taking the sample (like stirring them?).
Then I’d look at the probability of false alarm and missed alarm and adapt the sampling so as to set these to some required levels.I think there is a chance that there are no requirements stated in these terms – in which case discussing them might be very helpful in itself.

Regards
Sandor

0
#115081

Bob J
Participant

Tim,
Nicely summarized….
Best Regards,
Bob J

0
#115100

Aquinas
Member

I think  some sort of test, maybe an ANOVA, would be in order here to determine if the size of the tank is a factor.

0
#115490

JC
Participant

Don’t reinvent the wheel!  Search the open literature and apply a standard test! ASTM?

0
#115496

BK
Participant

Part of what should be considered is related more to chemical engineering….1.  If any segmentation could occur, a representative sample from a tank would require recirculating the tank contents to achieve adequate mixing with recirc times being pre-determined (usually the equivalent of turning the volume over 3 times).  2.  If the contents of the tank are appropriately mixed, the way in which the sample is drawn becomes critical if the sample is taken via a sampling line, where line size and flush time come into play.

0
#115702

Bob A.
Participant

Howdy
I would assume the result being looked for is either presence or no presence of a microorganism.  A well mixed vs. stratified liquid complicates the issue but this has been covered in the previous replies.  If a homogenous mixture, all that can be determined from a single sample with a +/- test result is the likely limit of concentration of microorganisms per unit volume.  A larger sample decreases this limit but of course complicates the testing.  There are FDA, EPA and USP guidelines for test methods, detectable limits and interpretation of results.
There is an old book, “Quantitative Microbiology” (i think) that is very helpful.  I of course do not rememebr the author.  Sorry

0
Viewing 8 posts - 1 through 8 (of 8 total)

The forum ‘General’ is closed to new topics and replies.