View Single Post
  #6   Report Post  
Posted to microsoft.public.excel.programming
Tom Mort Tom Mort is offline
external usenet poster
 
Posts: 5
Default taking average of each set of all possible combinations


I thought I sent a reply yesterday but don't see it.

What I'm trying to do is set up a program to help companies make an
informed decesion as to how many wastewater sample to take for
determining extra user charges and to compare with a monthly average
limit.

Certain business like bakeries etc have stronger sewage than regular
households. To make up the difference in costs to treat this sewage,
cites normally charge extra for concentrations above a certain amount.
Sampling is required to determine the concentration.

Also, at least in our community, there will be individual upper
concentration limits (a monthly average) that companies are not to
exceed. This is to insure that the municipal wasteater plant treatment
capacity is not exceeded.

In the past just a few samples per quarter were taken to assess the
extra charge. As an upper limit comes into play businesses will be
urged to perform self monitoring to assess compliance. That leads to
the question of how frequently to do this.

Each individual business has concentrations within a renge unique to
that business. The upper limits (at least intially) are based on the
90th percentile or so of sampling results for the past couple years. On
the one hand if a business doesn't have much variation they might be
able to only take one sample per month and be reasonably sure they have
a representative sample and that this one sample would meet the average
limit. On the other extreme to account for variation a sample could be
taken every day. That way the most representative sample would be taken
and any high anomolies would be averaged out.

What I't like is to have something that the businesses could enter their
past results for a period of time and the number of samples they
proposed to take per month and it would tell them what the odds
(assuming current conditions and variability are within the range of
past) that the average concentraion from this many samples would be
under the average limit.


*** Sent via Developersdex http://www.developersdex.com ***
Don't just participate in USENET...get rewarded for it!