The second step consists of eliminating from
the CAM the rows relating to the alter-natives
with a COD percentage removal less than 40%,
thus obtaining a simplified alternative matrix
(SAM).
The third step is to normalize the SAM. The
normalization is obtained by dividing each value
in the first column (relating to COD removal) by
the maximum value of the same column, and then
dividing the minimum value in the other columns
by the considered value.
The fourth step is to attribute a weight to each
criteria, distributing the value 10 among the 5
criteria adopted. An “equilibrate” weight distribution is obtained by multiplying each value by
2. This corresponds to giving a 20% relevance to
each criteria in the choice of the best coagulant
and dose. Obviously, the same result could be
obtained by dividing the value 100 among the 5
adopted criteria. In this case, an “equilibrate”
weight distribution should be obtained by multiplying each value by 20. In general, if there is no
interest in giving a different weight to the several
criteria, the normalized value can be directly
considered as it is.
The fifth step is to calculate for each alternative an average index (AI) obtained by taking
the average of the 5 weighted values for each
alternative.
The sixth and final step is to list the alternatives in a decreasing order in terms of AI. The
first coagulant and the relative dose will be the
best choice.