I have collected a bunch of data from Summer Schools in Germany and I would like to know which one of this is the best Summer School based on specified criterias using MCDM analysis.
Similar cases from the author of this method T. Saati:
1) Saaty, Thomas L.; Peniwati, Kirti (2008). Group Decision Making: Drawing out and Reconciling Differences. Pittsburgh, Pennsylvania: RWS Publications. ISBN 978-1-888603-08-8.
2)Saaty, Thomas L. Decision Making for Leaders: The Analytic Hierarchy Process for Decisions in a Complex World. — Pittsburgh, Pennsylvania : RWS Publications, 1999-05-01.
3)Saaty, Thomas L. The Hierarchon: A Dictionary of Hierarchies. — Pittsburgh, Pennsylvania : RWS Publications, 1992. — ISBN 0-9620317-5-5. 496 pages, spiral bound. Each entry includes a description and diagram of an AHP model; the models are grouped in categories: educational, government/public policy, government public/strategy, health military, non-profit, personal, planning, political, etc.
Similar cases from the author of this method T. Saati:
1) Saaty, Thomas L.; Peniwati, Kirti (2008). Group Decision Making: Drawing out and Reconciling Differences. Pittsburgh, Pennsylvania: RWS Publications. ISBN 978-1-888603-08-8.
2)Saaty, Thomas L. Decision Making for Leaders: The Analytic Hierarchy Process for Decisions in a Complex World. — Pittsburgh, Pennsylvania : RWS Publications, 1999-05-01.
3)Saaty, Thomas L. The Hierarchon: A Dictionary of Hierarchies. — Pittsburgh, Pennsylvania : RWS Publications, 1992. — ISBN 0-9620317-5-5. 496 pages, spiral bound. Each entry includes a description and diagram of an AHP model; the models are grouped in categories: educational, government/public policy, government public/strategy, health military, non-profit, personal, planning, political, etc.
The simplest solution to your problem is conducting some kind of "primitive" normalization:
I. Make new empty table 54 Alternatives and 6 criteria
II. Calculate the normalized table by some of several types of normalization - choose one:
1) Normalization 1:
a) for beneficial criteria - divide each value with the highest value in the column
b) for non beneficial criteria - divide minimal value in the column with each value
2) Normalization 2:
a) for beneficial criteria - calculate (1- minvalueofthecolumn/eachvalue)
b) for non beneficial criteria - calculate (1- eachvalue/highestvalueinthecolumn)
3) Normalization 0-1
a) for beneficial criteria - calculate (eachvalue-lovest)/(highest-lowest)
b) for non beneficial criteria - calculate (highestvalueofcolumn-each value)/(highest-lowest)
4) Euclidean normalisation
First - convert non beneficial criteria to beneficial - calculate the reciprocal value
Second - for each criterion, calculate Eucledian constant - it is a root of sum of squares of all values
Third - divide each value from the table with related Euclidean constant
III: After the normalized matrix is calculated, for each alternative multiply the criteria weight with normalized value, and sum those products for all criteria (SAW method - simple additive weighting).
The alternative with the highest sum of products is the winner.
The alternative can be conducting the Electre method - there is a free tool called Sanna (Excel addin for that). Also, the AHP is the alternative.
There are other types of normalization - normalization by sum, normalization by absolute value...
If you need additional help in Excel, let me know.
Applying several types of normalization and comparison of the results can be some sort of sensitive analysis.
Nikola Kadoić Could you please share citations for the "primitive normalization" that you have mentioned? I would definitely like to add this in my Thesis work!